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
,
91 PerRequestProxyHandler
,
99 register_socks_protocols
,
100 RejectedVideoReached
,
101 remove_terminal_sequences
,
115 supports_terminal_sequences
,
120 UnavailableVideoError
,
126 YoutubeDLCookieProcessor
,
128 YoutubeDLRedirectHandler
,
130 from .cache
import Cache
131 from .minicurses
import format_text
132 from .extractor
import (
133 gen_extractor_classes
,
136 _PLUGIN_CLASSES
as plugin_extractors
138 from .extractor
.openload
import PhantomJSwrapper
139 from .downloader
import (
141 get_suitable_downloader
,
142 shorten_protocol_name
144 from .downloader
.rtmp
import rtmpdump_version
145 from .postprocessor
import (
148 FFmpegFixupDuplicateMoovPP
,
149 FFmpegFixupDurationPP
,
152 FFmpegFixupStretchedPP
,
153 FFmpegFixupTimestampPP
,
156 MoveFilesAfterDownloadPP
,
157 _PLUGIN_CLASSES
as plugin_postprocessors
159 from .update
import detect_variant
160 from .version
import __version__
, RELEASE_GIT_HEAD
162 if compat_os_name
== 'nt':
166 class YoutubeDL(object):
169 YoutubeDL objects are the ones responsible of downloading the
170 actual video file and writing it to disk if the user has requested
171 it, among some other tasks. In most cases there should be one per
172 program. As, given a video URL, the downloader doesn't know how to
173 extract all the needed information, task that InfoExtractors do, it
174 has to pass the URL to one of them.
176 For this, YoutubeDL objects have a method that allows
177 InfoExtractors to be registered in a given order. When it is passed
178 a URL, the YoutubeDL object handles it to the first InfoExtractor it
179 finds that reports being able to handle it. The InfoExtractor extracts
180 all the information about the video or videos the URL refers to, and
181 YoutubeDL process the extracted information, possibly using a File
182 Downloader to download the video.
184 YoutubeDL objects accept a lot of parameters. In order not to saturate
185 the object constructor with arguments, it receives a dictionary of
186 options instead. These options are available through the params
187 attribute for the InfoExtractors to use. The YoutubeDL also
188 registers itself as the downloader in charge for the InfoExtractors
189 that are added to it, so this is a "mutual registration".
193 username: Username for authentication purposes.
194 password: Password for authentication purposes.
195 videopassword: Password for accessing a video.
196 ap_mso: Adobe Pass multiple-system operator identifier.
197 ap_username: Multiple-system operator account username.
198 ap_password: Multiple-system operator account password.
199 usenetrc: Use netrc for authentication instead.
200 verbose: Print additional info to stdout.
201 quiet: Do not print messages to stdout.
202 no_warnings: Do not print out anything for warnings.
203 forceprint: A dict with keys video/playlist mapped to
204 a list of templates to force print to stdout
205 For compatibility, a single list is also accepted
206 forceurl: Force printing final URL. (Deprecated)
207 forcetitle: Force printing title. (Deprecated)
208 forceid: Force printing ID. (Deprecated)
209 forcethumbnail: Force printing thumbnail URL. (Deprecated)
210 forcedescription: Force printing description. (Deprecated)
211 forcefilename: Force printing final filename. (Deprecated)
212 forceduration: Force printing duration. (Deprecated)
213 forcejson: Force printing info_dict as JSON.
214 dump_single_json: Force printing the info_dict of the whole playlist
215 (or video) as a single JSON line.
216 force_write_download_archive: Force writing download archive regardless
217 of 'skip_download' or 'simulate'.
218 simulate: Do not download the video files. If unset (or None),
219 simulate only if listsubtitles, listformats or list_thumbnails is used
220 format: Video format code. see "FORMAT SELECTION" for more details.
221 You can also pass a function. The function takes 'ctx' as
222 argument and returns the formats to download.
223 See "build_format_selector" for an implementation
224 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
225 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
226 extracting metadata even if the video is not actually
227 available for download (experimental)
228 format_sort: A list of fields by which to sort the video formats.
229 See "Sorting Formats" for more details.
230 format_sort_force: Force the given format_sort. see "Sorting Formats"
232 allow_multiple_video_streams: Allow multiple video streams to be merged
234 allow_multiple_audio_streams: Allow multiple audio streams to be merged
236 check_formats Whether to test if the formats are downloadable.
237 Can be True (check all), False (check none),
238 'selected' (check selected formats),
239 or None (check only if requested by extractor)
240 paths: Dictionary of output paths. The allowed keys are 'home'
241 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
242 outtmpl: Dictionary of templates for output names. Allowed keys
243 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
244 For compatibility with youtube-dl, a single string can also be used
245 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
246 restrictfilenames: Do not allow "&" and spaces in file names
247 trim_file_name: Limit length of filename (extension excluded)
248 windowsfilenames: Force the filenames to be windows compatible
249 ignoreerrors: Do not stop on download/postprocessing errors.
250 Can be 'only_download' to ignore only download errors.
251 Default is 'only_download' for CLI, but False for API
252 skip_playlist_after_errors: Number of allowed failures until the rest of
253 the playlist is skipped
254 force_generic_extractor: Force downloader to use the generic extractor
255 overwrites: Overwrite all video and metadata files if True,
256 overwrite only non-video files if None
257 and don't overwrite any file if False
258 For compatibility with youtube-dl,
259 "nooverwrites" may also be used instead
260 playliststart: Playlist item to start at.
261 playlistend: Playlist item to end at.
262 playlist_items: Specific indices of playlist to download.
263 playlistreverse: Download playlist items in reverse order.
264 playlistrandom: Download playlist items in random order.
265 matchtitle: Download only matching titles.
266 rejecttitle: Reject downloads for matching titles.
267 logger: Log messages to a logging.Logger instance.
268 logtostderr: Log messages to stderr instead of stdout.
269 consoletitle: Display progress in console window's titlebar.
270 writedescription: Write the video description to a .description file
271 writeinfojson: Write the video description to a .info.json file
272 clean_infojson: Remove private fields from the infojson
273 getcomments: Extract video comments. This will not be written to disk
274 unless writeinfojson is also given
275 writeannotations: Write the video annotations to a .annotations.xml file
276 writethumbnail: Write the thumbnail image to a file
277 allow_playlist_files: Whether to write playlists' description, infojson etc
278 also to disk when using the 'write*' options
279 write_all_thumbnails: Write all thumbnail formats to files
280 writelink: Write an internet shortcut file, depending on the
281 current platform (.url/.webloc/.desktop)
282 writeurllink: Write a Windows internet shortcut file (.url)
283 writewebloclink: Write a macOS internet shortcut file (.webloc)
284 writedesktoplink: Write a Linux internet shortcut file (.desktop)
285 writesubtitles: Write the video subtitles to a file
286 writeautomaticsub: Write the automatically generated subtitles to a file
287 allsubtitles: Deprecated - Use subtitleslangs = ['all']
288 Downloads all the subtitles of the video
289 (requires writesubtitles or writeautomaticsub)
290 listsubtitles: Lists all available subtitles for the video
291 subtitlesformat: The format code for subtitles
292 subtitleslangs: List of languages of the subtitles to download (can be regex).
293 The list may contain "all" to refer to all the available
294 subtitles. The language can be prefixed with a "-" to
295 exclude it from the requested languages. Eg: ['all', '-live_chat']
296 keepvideo: Keep the video file after post-processing
297 daterange: A DateRange object, download only if the upload_date is in the range.
298 skip_download: Skip the actual download of the video file
299 cachedir: Location of the cache files in the filesystem.
300 False to disable filesystem cache.
301 noplaylist: Download single video instead of a playlist if in doubt.
302 age_limit: An integer representing the user's age in years.
303 Unsuitable videos for the given age are skipped.
304 min_views: An integer representing the minimum view count the video
305 must have in order to not be skipped.
306 Videos without view count information are always
307 downloaded. None for no limit.
308 max_views: An integer representing the maximum view count.
309 Videos that are more popular than that are not
311 Videos without view count information are always
312 downloaded. None for no limit.
313 download_archive: File name of a file where all downloads are recorded.
314 Videos already present in the file are not downloaded
316 break_on_existing: Stop the download process after attempting to download a
317 file that is in the archive.
318 break_on_reject: Stop the download process when encountering a video that
319 has been filtered out.
320 break_per_url: Whether break_on_reject and break_on_existing
321 should act on each input URL as opposed to for the entire queue
322 cookiefile: File name where cookies should be read from and dumped to
323 cookiesfrombrowser: A tuple containing the name of the browser, the profile
324 name/pathfrom where cookies are loaded, and the name of the
325 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
326 nocheckcertificate: Do not verify SSL certificates
327 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
328 At the moment, this is only supported by YouTube.
329 proxy: URL of the proxy server to use
330 geo_verification_proxy: URL of the proxy to use for IP address verification
331 on geo-restricted sites.
332 socket_timeout: Time to wait for unresponsive hosts, in seconds
333 bidi_workaround: Work around buggy terminals without bidirectional text
334 support, using fridibi
335 debug_printtraffic:Print out sent and received HTTP traffic
336 include_ads: Download ads as well (deprecated)
337 default_search: Prepend this string if an input url is not valid.
338 'auto' for elaborate guessing
339 encoding: Use this encoding instead of the system-specified.
340 extract_flat: Do not resolve URLs, return the immediate result.
341 Pass in 'in_playlist' to only show this behavior for
343 wait_for_video: If given, wait for scheduled streams to become available.
344 The value should be a tuple containing the range
345 (min_secs, max_secs) to wait between retries
346 postprocessors: A list of dictionaries, each with an entry
347 * key: The name of the postprocessor. See
348 yt_dlp/postprocessor/__init__.py for a list.
349 * when: When to run the postprocessor. Can be one of
350 pre_process|before_dl|post_process|after_move.
351 Assumed to be 'post_process' if not given
352 post_hooks: Deprecated - Register a custom postprocessor instead
353 A list of functions that get called as the final step
354 for each video file, after all postprocessors have been
355 called. The filename will be passed as the only argument.
356 progress_hooks: A list of functions that get called on download
357 progress, with a dictionary with the entries
358 * status: One of "downloading", "error", or "finished".
359 Check this first and ignore unknown values.
360 * info_dict: The extracted info_dict
362 If status is one of "downloading", or "finished", the
363 following properties may also be present:
364 * filename: The final filename (always present)
365 * tmpfilename: The filename we're currently writing to
366 * downloaded_bytes: Bytes on disk
367 * total_bytes: Size of the whole file, None if unknown
368 * total_bytes_estimate: Guess of the eventual file size,
370 * elapsed: The number of seconds since download started.
371 * eta: The estimated time in seconds, None if unknown
372 * speed: The download speed in bytes/second, None if
374 * fragment_index: The counter of the currently
375 downloaded video fragment.
376 * fragment_count: The number of fragments (= individual
377 files that will be merged)
379 Progress hooks are guaranteed to be called at least once
380 (with status "finished") if the download is successful.
381 postprocessor_hooks: A list of functions that get called on postprocessing
382 progress, with a dictionary with the entries
383 * status: One of "started", "processing", or "finished".
384 Check this first and ignore unknown values.
385 * postprocessor: Name of the postprocessor
386 * info_dict: The extracted info_dict
388 Progress hooks are guaranteed to be called at least twice
389 (with status "started" and "finished") if the processing is successful.
390 merge_output_format: Extension to use when merging formats.
391 final_ext: Expected final extension; used to detect when the file was
392 already downloaded and converted
393 fixup: Automatically correct known faults of the file.
395 - "never": do nothing
396 - "warn": only emit a warning
397 - "detect_or_warn": check whether we can do anything
398 about it, warn otherwise (default)
399 source_address: Client-side IP address to bind to.
400 call_home: Boolean, true iff we are allowed to contact the
401 yt-dlp servers for debugging. (BROKEN)
402 sleep_interval_requests: Number of seconds to sleep between requests
404 sleep_interval: Number of seconds to sleep before each download when
405 used alone or a lower bound of a range for randomized
406 sleep before each download (minimum possible number
407 of seconds to sleep) when used along with
409 max_sleep_interval:Upper bound of a range for randomized sleep before each
410 download (maximum possible number of seconds to sleep).
411 Must only be used along with sleep_interval.
412 Actual sleep time will be a random float from range
413 [sleep_interval; max_sleep_interval].
414 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
415 listformats: Print an overview of available video formats and exit.
416 list_thumbnails: Print a table of all thumbnails and exit.
417 match_filter: A function that gets called with the info_dict of
419 If it returns a message, the video is ignored.
420 If it returns None, the video is downloaded.
421 match_filter_func in utils.py is one example for this.
422 no_color: Do not emit color codes in output.
423 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
426 Two-letter ISO 3166-2 country code that will be used for
427 explicit geographic restriction bypassing via faking
428 X-Forwarded-For HTTP header
430 IP range in CIDR notation that will be used similarly to
433 The following options determine which downloader is picked:
434 external_downloader: A dictionary of protocol keys and the executable of the
435 external downloader to use for it. The allowed protocols
436 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
437 Set the value to 'native' to use the native downloader
438 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
439 or {'m3u8': 'ffmpeg'} instead.
440 Use the native HLS downloader instead of ffmpeg/avconv
441 if True, otherwise use ffmpeg/avconv if False, otherwise
442 use downloader suggested by extractor if None.
443 compat_opts: Compatibility options. See "Differences in default behavior".
444 The following options do not work when used through the API:
445 filename, abort-on-error, multistreams, no-live-chat, format-sort
446 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
447 Refer __init__.py for their implementation
448 progress_template: Dictionary of templates for progress outputs.
449 Allowed keys are 'download', 'postprocess',
450 'download-title' (console title) and 'postprocess-title'.
451 The template is mapped on a dictionary with keys 'progress' and 'info'
453 The following parameters are not used by YoutubeDL itself, they are used by
454 the downloader (see yt_dlp/downloader/common.py):
455 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
456 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
457 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
458 external_downloader_args, concurrent_fragment_downloads.
460 The following options are used by the post processors:
461 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
462 otherwise prefer ffmpeg. (avconv support is deprecated)
463 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
464 to the binary or its containing directory.
465 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
466 and a list of additional command-line arguments for the
467 postprocessor/executable. The dict can also have "PP+EXE" keys
468 which are used when the given exe is used by the given PP.
469 Use 'default' as the name for arguments to passed to all PP
470 For compatibility with youtube-dl, a single list of args
473 The following options are used by the extractors:
474 extractor_retries: Number of times to retry for known errors
475 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
476 hls_split_discontinuity: Split HLS playlists to different formats at
477 discontinuities such as ad breaks (default: False)
478 extractor_args: A dictionary of arguments to be passed to the extractors.
479 See "EXTRACTOR ARGUMENTS" for details.
480 Eg: {'youtube': {'skip': ['dash', 'hls']}}
481 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
482 If True (default), DASH manifests and related
483 data will be downloaded and processed by extractor.
484 You can reduce network I/O by disabling it if you don't
485 care about DASH. (only for youtube)
486 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
487 If True (default), HLS manifests and related
488 data will be downloaded and processed by extractor.
489 You can reduce network I/O by disabling it if you don't
490 care about HLS. (only for youtube)
493 _NUMERIC_FIELDS
= set((
494 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
495 'timestamp', 'release_timestamp',
496 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
497 'average_rating', 'comment_count', 'age_limit',
498 'start_time', 'end_time',
499 'chapter_number', 'season_number', 'episode_number',
500 'track_number', 'disc_number', 'release_year',
503 _format_selection_exts
= {
504 'audio': {'m4a', 'mp3', 'ogg', 'aac'}
,
505 'video': {'mp4', 'flv', 'webm', '3gp'}
,
506 'storyboards': {'mhtml'}
,
511 _pps
= {k: [] for k in POSTPROCESS_WHEN}
512 _printed_messages
= set()
513 _first_webpage_request
= True
514 _download_retcode
= None
515 _num_downloads
= None
517 _playlist_urls
= set()
520 def __init__(self
, params
=None, auto_init
=True):
521 """Create a FileDownloader object with the given options.
522 @param auto_init Whether to load the default extractors and print header (if verbose).
523 Set to 'no_verbose_header' to not print the header
528 self
._ies
_instances
= {}
529 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
530 self
._printed
_messages
= set()
531 self
._first
_webpage
_request
= True
532 self
._post
_hooks
= []
533 self
._progress
_hooks
= []
534 self
._postprocessor
_hooks
= []
535 self
._download
_retcode
= 0
536 self
._num
_downloads
= 0
538 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
539 self
._err
_file
= sys
.stderr
541 self
.cache
= Cache(self
)
543 windows_enable_vt_mode()
544 self
._allow
_colors
= {
545 'screen': not self
.params
.get('no_color') and supports_terminal_sequences(self
._screen
_file
),
546 'err': not self
.params
.get('no_color') and supports_terminal_sequences(self
._err
_file
),
549 if sys
.version_info
< (3, 6):
551 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
553 if self
.params
.get('allow_unplayable_formats'):
555 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
556 'This is a developer option intended for debugging. \n'
557 ' If you experience any issues while using this option, '
558 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
560 def check_deprecated(param
, option
, suggestion
):
561 if self
.params
.get(param
) is not None:
562 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
566 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
567 if self
.params
.get('geo_verification_proxy') is None:
568 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
570 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
571 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
572 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
574 for msg
in self
.params
.get('_warnings', []):
575 self
.report_warning(msg
)
576 for msg
in self
.params
.get('_deprecation_warnings', []):
577 self
.deprecation_warning(msg
)
579 if 'list-formats' in self
.params
.get('compat_opts', []):
580 self
.params
['listformats_table'] = False
582 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
583 # nooverwrites was unnecessarily changed to overwrites
584 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
585 # This ensures compatibility with both keys
586 self
.params
['overwrites'] = not self
.params
['nooverwrites']
587 elif self
.params
.get('overwrites') is None:
588 self
.params
.pop('overwrites', None)
590 self
.params
['nooverwrites'] = not self
.params
['overwrites']
592 # Compatibility with older syntax
593 params
.setdefault('forceprint', {})
594 if not isinstance(params
['forceprint'], dict):
595 params
['forceprint'] = {'video': params['forceprint']}
597 if params
.get('bidi_workaround', False):
600 master
, slave
= pty
.openpty()
601 width
= compat_get_terminal_size().columns
605 width_args
= ['-w', str(width
)]
607 stdin
=subprocess
.PIPE
,
609 stderr
=self
._err
_file
)
611 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
613 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
614 self
._output
_channel
= os
.fdopen(master
, 'rb')
615 except OSError as ose
:
616 if ose
.errno
== errno
.ENOENT
:
618 'Could not find fribidi executable, ignoring --bidi-workaround. '
619 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
623 if (sys
.platform
!= 'win32'
624 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
625 and not params
.get('restrictfilenames', False)):
626 # Unicode filesystem API will throw errors (#1474, #13027)
628 'Assuming --restrict-filenames since file system encoding '
629 'cannot encode all characters. '
630 'Set the LC_ALL environment variable to fix this.')
631 self
.params
['restrictfilenames'] = True
633 self
.outtmpl_dict
= self
.parse_outtmpl()
635 # Creating format selector here allows us to catch syntax errors before the extraction
636 self
.format_selector
= (
637 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
638 else self
.params
['format'] if callable(self
.params
['format'])
639 else self
.build_format_selector(self
.params
['format']))
644 if auto_init
!= 'no_verbose_header':
645 self
.print_debug_header()
646 self
.add_default_info_extractors()
649 'post_hooks': self
.add_post_hook
,
650 'progress_hooks': self
.add_progress_hook
,
651 'postprocessor_hooks': self
.add_postprocessor_hook
,
653 for opt
, fn
in hooks
.items():
654 for ph
in self
.params
.get(opt
, []):
657 for pp_def_raw
in self
.params
.get('postprocessors', []):
658 pp_def
= dict(pp_def_raw
)
659 when
= pp_def
.pop('when', 'post_process')
660 self
.add_post_processor(
661 get_postprocessor(pp_def
.pop('key'))(self
, **compat_kwargs(pp_def
)),
664 register_socks_protocols()
666 def preload_download_archive(fn
):
667 """Preload the archive, if any is specified"""
670 self
.write_debug(f
'Loading archive file {fn!r}')
672 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
673 for line
in archive_file
:
674 self
.archive
.add(line
.strip())
675 except IOError as ioe
:
676 if ioe
.errno
!= errno
.ENOENT
:
682 preload_download_archive(self
.params
.get('download_archive'))
684 def warn_if_short_id(self
, argv
):
685 # short YouTube ID starting with dash?
687 i
for i
, a
in enumerate(argv
)
688 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
692 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
693 + ['--'] + [argv
[i
] for i
in idxs
]
696 'Long argument string detected. '
697 'Use -- to separate parameters and URLs, like this:\n%s' %
698 args_to_str(correct_argv
))
700 def add_info_extractor(self
, ie
):
701 """Add an InfoExtractor object to the end of the list."""
703 self
._ies
[ie_key
] = ie
704 if not isinstance(ie
, type):
705 self
._ies
_instances
[ie_key
] = ie
706 ie
.set_downloader(self
)
708 def _get_info_extractor_class(self
, ie_key
):
709 ie
= self
._ies
.get(ie_key
)
711 ie
= get_info_extractor(ie_key
)
712 self
.add_info_extractor(ie
)
715 def get_info_extractor(self
, ie_key
):
717 Get an instance of an IE with name ie_key, it will try to get one from
718 the _ies list, if there's no instance it will create a new one and add
719 it to the extractor list.
721 ie
= self
._ies
_instances
.get(ie_key
)
723 ie
= get_info_extractor(ie_key
)()
724 self
.add_info_extractor(ie
)
727 def add_default_info_extractors(self
):
729 Add the InfoExtractors returned by gen_extractors to the end of the list
731 for ie
in gen_extractor_classes():
732 self
.add_info_extractor(ie
)
734 def add_post_processor(self
, pp
, when
='post_process'):
735 """Add a PostProcessor object to the end of the chain."""
736 self
._pps
[when
].append(pp
)
737 pp
.set_downloader(self
)
739 def add_post_hook(self
, ph
):
740 """Add the post hook"""
741 self
._post
_hooks
.append(ph
)
743 def add_progress_hook(self
, ph
):
744 """Add the download progress hook"""
745 self
._progress
_hooks
.append(ph
)
747 def add_postprocessor_hook(self
, ph
):
748 """Add the postprocessing progress hook"""
749 self
._postprocessor
_hooks
.append(ph
)
750 for pps
in self
._pps
.values():
752 pp
.add_progress_hook(ph
)
754 def _bidi_workaround(self
, message
):
755 if not hasattr(self
, '_output_channel'):
758 assert hasattr(self
, '_output_process')
759 assert isinstance(message
, compat_str
)
760 line_count
= message
.count('\n') + 1
761 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
762 self
._output
_process
.stdin
.flush()
763 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
764 for _
in range(line_count
))
765 return res
[:-len('\n')]
767 def _write_string(self
, message
, out
=None, only_once
=False):
769 if message
in self
._printed
_messages
:
771 self
._printed
_messages
.add(message
)
772 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
774 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
775 """Print message to stdout"""
776 if self
.params
.get('logger'):
777 self
.params
['logger'].debug(message
)
778 elif not quiet
or self
.params
.get('verbose'):
780 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
781 self
._err
_file
if quiet
else self
._screen
_file
)
783 def to_stderr(self
, message
, only_once
=False):
784 """Print message to stderr"""
785 assert isinstance(message
, compat_str
)
786 if self
.params
.get('logger'):
787 self
.params
['logger'].error(message
)
789 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._err
_file
, only_once
=only_once
)
791 def to_console_title(self
, message
):
792 if not self
.params
.get('consoletitle', False):
794 message
= remove_terminal_sequences(message
)
795 if compat_os_name
== 'nt':
796 if ctypes
.windll
.kernel32
.GetConsoleWindow():
797 # c_wchar_p() might not be necessary if `message` is
798 # already of type unicode()
799 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
800 elif 'TERM' in os
.environ
:
801 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
803 def save_console_title(self
):
804 if not self
.params
.get('consoletitle', False):
806 if self
.params
.get('simulate'):
808 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
809 # Save the title on stack
810 self
._write
_string
('\033[22;0t', self
._screen
_file
)
812 def restore_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 # Restore the title from stack
819 self
._write
_string
('\033[23;0t', self
._screen
_file
)
822 self
.save_console_title()
825 def __exit__(self
, *args
):
826 self
.restore_console_title()
828 if self
.params
.get('cookiefile') is not None:
829 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
831 def trouble(self
, message
=None, tb
=None, is_error
=True):
832 """Determine action to take when a download problem appears.
834 Depending on if the downloader has been configured to ignore
835 download errors or not, this method may throw an exception or
836 not when errors are found, after printing the message.
838 @param tb If given, is additional traceback information
839 @param is_error Whether to raise error according to ignorerrors
841 if message
is not None:
842 self
.to_stderr(message
)
843 if self
.params
.get('verbose'):
845 if sys
.exc_info()[0]: # if .trouble has been called from an except block
847 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
848 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
849 tb
+= encode_compat_str(traceback
.format_exc())
851 tb_data
= traceback
.format_list(traceback
.extract_stack())
852 tb
= ''.join(tb_data
)
857 if not self
.params
.get('ignoreerrors'):
858 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
859 exc_info
= sys
.exc_info()[1].exc_info
861 exc_info
= sys
.exc_info()
862 raise DownloadError(message
, exc_info
)
863 self
._download
_retcode
= 1
865 def to_screen(self
, message
, skip_eol
=False):
866 """Print message to stdout if not in quiet mode"""
868 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
872 EMPHASIS
= 'light blue'
877 SUPPRESS
= 'light black'
879 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
882 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', 'ascii')
883 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
884 if fallback
is not None and text
!= original_text
:
886 if isinstance(f
, self
.Styles
):
888 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
890 def _format_screen(self
, *args
, **kwargs
):
891 return self
._format
_text
(
892 self
._screen
_file
, self
._allow
_colors
['screen'], *args
, **kwargs
)
894 def _format_err(self
, *args
, **kwargs
):
895 return self
._format
_text
(
896 self
._err
_file
, self
._allow
_colors
['err'], *args
, **kwargs
)
898 def report_warning(self
, message
, only_once
=False):
900 Print the message to stderr, it will be prefixed with 'WARNING:'
901 If stderr is a tty file the 'WARNING:' will be colored
903 if self
.params
.get('logger') is not None:
904 self
.params
['logger'].warning(message
)
906 if self
.params
.get('no_warnings'):
908 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
910 def deprecation_warning(self
, message
):
911 if self
.params
.get('logger') is not None:
912 self
.params
['logger'].warning('DeprecationWarning: {message}')
914 self
.to_stderr(f
'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
916 def report_error(self
, message
, *args
, **kwargs
):
918 Do the same as trouble, but prefixes the message with 'ERROR:', colored
919 in red if stderr is a tty file.
921 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
923 def write_debug(self
, message
, only_once
=False):
924 '''Log debug message or Print message to stderr'''
925 if not self
.params
.get('verbose', False):
927 message
= '[debug] %s' % message
928 if self
.params
.get('logger'):
929 self
.params
['logger'].debug(message
)
931 self
.to_stderr(message
, only_once
)
933 def report_file_already_downloaded(self
, file_name
):
934 """Report file has already been fully downloaded."""
936 self
.to_screen('[download] %s has already been downloaded' % file_name
)
937 except UnicodeEncodeError:
938 self
.to_screen('[download] The file has already been downloaded')
940 def report_file_delete(self
, file_name
):
941 """Report that existing file will be deleted."""
943 self
.to_screen('Deleting existing file %s' % file_name
)
944 except UnicodeEncodeError:
945 self
.to_screen('Deleting existing file')
947 def raise_no_formats(self
, info
, forced
=False):
948 has_drm
= info
.get('__has_drm')
949 msg
= 'This video is DRM protected' if has_drm
else 'No video formats found!'
950 expected
= self
.params
.get('ignore_no_formats_error')
951 if forced
or not expected
:
952 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
953 expected
=has_drm
or expected
)
955 self
.report_warning(msg
)
957 def parse_outtmpl(self
):
958 outtmpl_dict
= self
.params
.get('outtmpl', {})
959 if not isinstance(outtmpl_dict
, dict):
960 outtmpl_dict
= {'default': outtmpl_dict}
961 # Remove spaces in the default template
962 if self
.params
.get('restrictfilenames'):
963 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
965 sanitize
= lambda x
: x
966 outtmpl_dict
.update({
967 k
: sanitize(v
) for k
, v
in DEFAULT_OUTTMPL
.items()
968 if outtmpl_dict
.get(k
) is None})
969 for key
, val
in outtmpl_dict
.items():
970 if isinstance(val
, bytes):
972 'Parameter outtmpl is bytes, but should be a unicode string. '
973 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
976 def get_output_path(self
, dir_type
='', filename
=None):
977 paths
= self
.params
.get('paths', {})
978 assert isinstance(paths
, dict)
980 expand_path(paths
.get('home', '').strip()),
981 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
984 # Temporary fix for #4787
985 # 'Treat' all problem characters by passing filename through preferredencoding
986 # to workaround encoding issues with subprocess on python2 @ Windows
987 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
988 path
= encodeFilename(path
, True).decode(preferredencoding())
989 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
992 def _outtmpl_expandpath(outtmpl
):
993 # expand_path translates '%%' into '%' and '$$' into '$'
994 # correspondingly that is not what we want since we need to keep
995 # '%%' intact for template dict substitution step. Working around
996 # with boundary-alike separator hack.
997 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
998 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
1000 # outtmpl should be expand_path'ed before template dict substitution
1001 # because meta fields may contain env variables we don't want to
1002 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1003 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1004 return expand_path(outtmpl
).replace(sep
, '')
1007 def escape_outtmpl(outtmpl
):
1008 ''' Escape any remaining strings like %s, %abc% etc. '''
1010 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1011 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1015 def validate_outtmpl(cls
, outtmpl
):
1016 ''' @return None or Exception object '''
1018 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljqBUDS]'),
1019 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1020 cls
._outtmpl
_expandpath
(outtmpl
))
1022 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1024 except ValueError as err
:
1028 def _copy_infodict(info_dict
):
1029 info_dict
= dict(info_dict
)
1030 for key
in ('__original_infodict', '__postprocessors'):
1031 info_dict
.pop(key
, None)
1034 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1035 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1036 @param sanitize Whether to sanitize the output as a filename.
1037 For backward compatibility, a function can also be passed
1040 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1042 info_dict
= self
._copy
_infodict
(info_dict
)
1043 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1044 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1045 if info_dict
.get('duration', None) is not None
1047 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
1048 info_dict
['video_autonumber'] = self
._num
_videos
1049 if info_dict
.get('resolution') is None:
1050 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1052 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1053 # of %(field)s to %(field)0Nd for backward compatibility
1054 field_size_compat_map
= {
1055 'playlist_index': number_of_digits(info_dict
.get('_last_playlist_index') or 0),
1056 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1057 'autonumber': self
.params
.get('autonumber_size') or 5,
1061 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljqBUDS]'))
1066 # Field is of the form key1.key2...
1067 # where keys (except first) can be string, int or slice
1068 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1069 MATH_FIELD_RE
= r
'''(?:{field}|{num})'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
1070 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1071 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
1074 (?P<maths>(?:{math_op}{math_field})*)
1075 (?:>(?P<strf_format>.+?))?
1076 (?P<alternate>(?<!\\),[^|&)]+)?
1077 (?:&(?P<replacement>.*?))?
1078 (?:\|(?P<default>.*?))?
1079 $'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
1081 def _traverse_infodict(k
):
1085 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1087 def get_value(mdict
):
1089 value
= _traverse_infodict(mdict
['fields'])
1092 value
= float_or_none(value
)
1093 if value
is not None:
1096 offset_key
= mdict
['maths']
1098 value
= float_or_none(value
)
1102 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1103 offset_key
).group(0)
1104 offset_key
= offset_key
[len(item
):]
1105 if operator
is None:
1106 operator
= MATH_FUNCTIONS
[item
]
1108 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1109 offset
= float_or_none(item
)
1111 offset
= float_or_none(_traverse_infodict(item
))
1113 value
= operator(value
, multiplier
* offset
)
1114 except (TypeError, ZeroDivisionError):
1117 # Datetime formatting
1118 if mdict
['strf_format']:
1119 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1123 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1125 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1126 return sanitize_filename(str(value
), restricted
=restricted
,
1127 is_id
=re
.search(r
'(^|[_.])id(\.|$)', key
))
1129 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1130 sanitize
= bool(sanitize
)
1132 def _dumpjson_default(obj
):
1133 if isinstance(obj
, (set, LazyList
)):
1137 def create_key(outer_mobj
):
1138 if not outer_mobj
.group('has_key'):
1139 return outer_mobj
.group(0)
1140 key
= outer_mobj
.group('key')
1141 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1142 initial_field
= mobj
.group('fields') if mobj
else ''
1143 value
, replacement
, default
= None, None, na
1145 mobj
= mobj
.groupdict()
1146 default
= mobj
['default'] if mobj
['default'] is not None else default
1147 value
= get_value(mobj
)
1148 replacement
= mobj
['replacement']
1149 if value
is None and mobj
['alternate']:
1150 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['alternate'][1:])
1154 fmt
= outer_mobj
.group('format')
1155 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1156 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
1158 value
= default
if value
is None else value
if replacement
is None else replacement
1160 flags
= outer_mobj
.group('conversion') or ''
1161 str_fmt
= f
'{fmt[:-1]}s'
1162 if fmt
[-1] == 'l': # list
1163 delim
= '\n' if '#' in flags
else ', '
1164 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1165 elif fmt
[-1] == 'j': # json
1166 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1167 elif fmt
[-1] == 'q': # quoted
1168 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1169 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1170 elif fmt
[-1] == 'B': # bytes
1171 value
= f
'%{str_fmt}'.encode('utf-8') % str(value
).encode('utf-8')
1172 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1173 elif fmt
[-1] == 'U': # unicode normalized
1174 value
, fmt
= unicodedata
.normalize(
1175 # "+" = compatibility equivalence, "#" = NFD
1176 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1178 elif fmt
[-1] == 'D': # decimal suffix
1179 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1180 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1181 factor
=1024 if '#' in flags
else 1000)
1182 elif fmt
[-1] == 'S': # filename sanitization
1183 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1184 elif fmt
[-1] == 'c':
1186 value
= str(value
)[0]
1189 elif fmt
[-1] not in 'rs': # numeric
1190 value
= float_or_none(value
)
1192 value
, fmt
= default
, 's'
1196 # If value is an object, sanitize might convert it to a string
1197 # So we convert it to repr first
1198 value
, fmt
= repr(value
), str_fmt
1199 if fmt
[-1] in 'csr':
1200 value
= sanitizer(initial_field
, value
)
1202 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1203 TMPL_DICT
[key
] = value
1204 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1206 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1208 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1209 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1210 return self
.escape_outtmpl(outtmpl
) % info_dict
1212 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
1214 outtmpl
= self
._outtmpl
_expandpath
(self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default']))
1215 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1217 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
1218 if filename
and force_ext
is not None:
1219 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1221 # https://github.com/blackjack4494/youtube-dlc/issues/85
1222 trim_file_name
= self
.params
.get('trim_file_name', False)
1224 no_ext
, *ext
= filename
.rsplit('.', 2)
1225 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1228 except ValueError as err
:
1229 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1232 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
1233 """Generate the output filename."""
1235 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
1236 if not filename
and dir_type
not in ('', 'temp'):
1240 if not self
.params
.get('paths'):
1242 elif filename
== '-':
1243 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1244 elif os
.path
.isabs(filename
):
1245 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1246 if filename
== '-' or not filename
:
1249 return self
.get_output_path(dir_type
, filename
)
1251 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1252 """ Returns None if the file should be downloaded """
1254 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1257 if 'title' in info_dict
:
1258 # This can happen when we're just evaluating the playlist
1259 title
= info_dict
['title']
1260 matchtitle
= self
.params
.get('matchtitle', False)
1262 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1263 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1264 rejecttitle
= self
.params
.get('rejecttitle', False)
1266 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1267 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1268 date
= info_dict
.get('upload_date')
1269 if date
is not None:
1270 dateRange
= self
.params
.get('daterange', DateRange())
1271 if date
not in dateRange
:
1272 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1273 view_count
= info_dict
.get('view_count')
1274 if view_count
is not None:
1275 min_views
= self
.params
.get('min_views')
1276 if min_views
is not None and view_count
< min_views
:
1277 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1278 max_views
= self
.params
.get('max_views')
1279 if max_views
is not None and view_count
> max_views
:
1280 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1281 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1282 return 'Skipping "%s" because it is age restricted' % video_title
1284 match_filter
= self
.params
.get('match_filter')
1285 if match_filter
is not None:
1287 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1289 # For backward compatibility
1290 ret
= None if incomplete
else match_filter(info_dict
)
1295 if self
.in_download_archive(info_dict
):
1296 reason
= '%s has already been recorded in the archive' % video_title
1297 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1299 reason
= check_filter()
1300 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1301 if reason
is not None:
1303 self
.to_screen('[download] ' + reason
)
1304 if self
.params
.get(break_opt
, False):
1309 def add_extra_info(info_dict
, extra_info
):
1310 '''Set the keys from extra_info in info dict if they are missing'''
1311 for key
, value
in extra_info
.items():
1312 info_dict
.setdefault(key
, value
)
1314 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1315 process
=True, force_generic_extractor
=False):
1317 Return a list with a dictionary for each video extracted.
1320 url -- URL to extract
1323 download -- whether to download videos during extraction
1324 ie_key -- extractor key hint
1325 extra_info -- dictionary containing the extra values to add to each result
1326 process -- whether to resolve all unresolved references (URLs, playlist items),
1327 must be True for download to work.
1328 force_generic_extractor -- force using the generic extractor
1331 if extra_info
is None:
1334 if not ie_key
and force_generic_extractor
:
1338 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1342 for ie_key
, ie
in ies
.items():
1343 if not ie
.suitable(url
):
1346 if not ie
.working():
1347 self
.report_warning('The program functionality for this site has been marked as broken, '
1348 'and will probably not work.')
1350 temp_id
= ie
.get_temp_id(url
)
1351 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1352 self
.to_screen(f
'[{ie_key}] {temp_id}: has already been recorded in the archive')
1353 if self
.params
.get('break_on_existing', False):
1354 raise ExistingVideoReached()
1356 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1358 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1360 def __handle_extraction_exceptions(func
):
1361 @functools.wraps(func
)
1362 def wrapper(self
, *args
, **kwargs
):
1365 return func(self
, *args
, **kwargs
)
1366 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1368 except ReExtractInfo
as e
:
1370 self
.to_screen(f
'{e}; Re-extracting data')
1372 self
.to_stderr('\r')
1373 self
.report_warning(f
'{e}; Re-extracting data')
1375 except GeoRestrictedError
as e
:
1378 msg
+= '\nThis video is available in %s.' % ', '.join(
1379 map(ISO3166Utils
.short2full
, e
.countries
))
1380 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1381 self
.report_error(msg
)
1382 except ExtractorError
as e
: # An error we somewhat expected
1383 self
.report_error(str(e
), e
.format_traceback())
1384 except Exception as e
:
1385 if self
.params
.get('ignoreerrors'):
1386 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1392 def _wait_for_video(self
, ie_result
):
1393 if (not self
.params
.get('wait_for_video')
1394 or ie_result
.get('_type', 'video') != 'video'
1395 or ie_result
.get('formats') or ie_result
.get('url')):
1398 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1403 self
.to_screen(msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r', skip_eol
=True)
1406 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1407 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1408 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1409 diff
= random
.randrange(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
)
1410 self
.report_warning('Release time of video is not known')
1411 elif (diff
or 0) <= 0:
1412 self
.report_warning('Video should already be available according to extracted info')
1413 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1414 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1416 wait_till
= time
.time() + diff
1419 diff
= wait_till
- time
.time()
1422 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1423 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1425 except KeyboardInterrupt:
1427 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1428 except BaseException
as e
:
1429 if not isinstance(e
, ReExtractInfo
):
1433 @__handle_extraction_exceptions
1434 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1435 ie_result
= ie
.extract(url
)
1436 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1438 if isinstance(ie_result
, list):
1439 # Backwards compatibility: old IE result format
1441 '_type': 'compat_list',
1442 'entries': ie_result
,
1444 if extra_info
.get('original_url'):
1445 ie_result
.setdefault('original_url', extra_info
['original_url'])
1446 self
.add_default_extra_info(ie_result
, ie
, url
)
1448 self
._wait
_for
_video
(ie_result
)
1449 return self
.process_ie_result(ie_result
, download
, extra_info
)
1453 def add_default_extra_info(self
, ie_result
, ie
, url
):
1455 self
.add_extra_info(ie_result
, {
1457 'original_url': url
,
1458 'webpage_url_basename': url_basename(url
),
1459 'webpage_url_domain': get_domain(url
),
1462 self
.add_extra_info(ie_result
, {
1463 'extractor': ie
.IE_NAME
,
1464 'extractor_key': ie
.ie_key(),
1467 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1469 Take the result of the ie(may be modified) and resolve all unresolved
1470 references (URLs, playlist items).
1472 It will also download the videos if 'download'.
1473 Returns the resolved ie_result.
1475 if extra_info
is None:
1477 result_type
= ie_result
.get('_type', 'video')
1479 if result_type
in ('url', 'url_transparent'):
1480 ie_result
['url'] = sanitize_url(ie_result
['url'])
1481 if ie_result
.get('original_url'):
1482 extra_info
.setdefault('original_url', ie_result
['original_url'])
1484 extract_flat
= self
.params
.get('extract_flat', False)
1485 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1486 or extract_flat
is True):
1487 info_copy
= ie_result
.copy()
1488 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1489 if ie
and not ie_result
.get('id'):
1490 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1491 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1492 self
.add_extra_info(info_copy
, extra_info
)
1493 info_copy
, _
= self
.pre_process(info_copy
)
1494 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1495 if self
.params
.get('force_write_download_archive', False):
1496 self
.record_download_archive(info_copy
)
1499 if result_type
== 'video':
1500 self
.add_extra_info(ie_result
, extra_info
)
1501 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1502 additional_urls
= (ie_result
or {}).get('additional_urls')
1504 # TODO: Improve MetadataParserPP to allow setting a list
1505 if isinstance(additional_urls
, compat_str
):
1506 additional_urls
= [additional_urls
]
1508 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1509 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1510 ie_result
['additional_entries'] = [
1512 url
, download
, extra_info
=extra_info
,
1513 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1514 for url
in additional_urls
1517 elif result_type
== 'url':
1518 # We have to add extra_info to the results because it may be
1519 # contained in a playlist
1520 return self
.extract_info(
1521 ie_result
['url'], download
,
1522 ie_key
=ie_result
.get('ie_key'),
1523 extra_info
=extra_info
)
1524 elif result_type
== 'url_transparent':
1525 # Use the information from the embedding page
1526 info
= self
.extract_info(
1527 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1528 extra_info
=extra_info
, download
=False, process
=False)
1530 # extract_info may return None when ignoreerrors is enabled and
1531 # extraction failed with an error, don't crash and return early
1536 force_properties
= dict(
1537 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1538 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1539 if f
in force_properties
:
1540 del force_properties
[f
]
1541 new_result
= info
.copy()
1542 new_result
.update(force_properties
)
1544 # Extracted info may not be a video result (i.e.
1545 # info.get('_type', 'video') != video) but rather an url or
1546 # url_transparent. In such cases outer metadata (from ie_result)
1547 # should be propagated to inner one (info). For this to happen
1548 # _type of info should be overridden with url_transparent. This
1549 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1550 if new_result
.get('_type') == 'url':
1551 new_result
['_type'] = 'url_transparent'
1553 return self
.process_ie_result(
1554 new_result
, download
=download
, extra_info
=extra_info
)
1555 elif result_type
in ('playlist', 'multi_video'):
1556 # Protect from infinite recursion due to recursively nested playlists
1557 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1558 webpage_url
= ie_result
['webpage_url']
1559 if webpage_url
in self
._playlist
_urls
:
1561 '[download] Skipping already downloaded playlist: %s'
1562 % ie_result
.get('title') or ie_result
.get('id'))
1565 self
._playlist
_level
+= 1
1566 self
._playlist
_urls
.add(webpage_url
)
1567 self
._sanitize
_thumbnails
(ie_result
)
1569 return self
.__process
_playlist
(ie_result
, download
)
1571 self
._playlist
_level
-= 1
1572 if not self
._playlist
_level
:
1573 self
._playlist
_urls
.clear()
1574 elif result_type
== 'compat_list':
1575 self
.report_warning(
1576 'Extractor %s returned a compat_list result. '
1577 'It needs to be updated.' % ie_result
.get('extractor'))
1580 self
.add_extra_info(r
, {
1581 'extractor': ie_result
['extractor'],
1582 'webpage_url': ie_result
['webpage_url'],
1583 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1584 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1585 'extractor_key': ie_result
['extractor_key'],
1588 ie_result
['entries'] = [
1589 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1590 for r
in ie_result
['entries']
1594 raise Exception('Invalid result type: %s' % result_type
)
1596 def _ensure_dir_exists(self
, path
):
1597 return make_dir(path
, self
.report_error
)
1600 def _playlist_infodict(ie_result
, **kwargs
):
1603 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1604 'playlist_id': ie_result
.get('id'),
1605 'playlist_title': ie_result
.get('title'),
1606 'playlist_uploader': ie_result
.get('uploader'),
1607 'playlist_uploader_id': ie_result
.get('uploader_id'),
1608 'playlist_index': 0,
1612 def __process_playlist(self
, ie_result
, download
):
1613 # We process each entry in the playlist
1614 playlist
= ie_result
.get('title') or ie_result
.get('id')
1615 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1617 if 'entries' not in ie_result
:
1618 raise EntryNotInPlaylist('There are no entries')
1620 MissingEntry
= object()
1621 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1622 if incomplete_entries
:
1623 def fill_missing_entries(entries
, indices
):
1624 ret
= [MissingEntry
] * max(indices
)
1625 for i
, entry
in zip(indices
, entries
):
1628 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1630 playlist_results
= []
1632 playliststart
= self
.params
.get('playliststart', 1)
1633 playlistend
= self
.params
.get('playlistend')
1634 # For backwards compatibility, interpret -1 as whole list
1635 if playlistend
== -1:
1638 playlistitems_str
= self
.params
.get('playlist_items')
1639 playlistitems
= None
1640 if playlistitems_str
is not None:
1641 def iter_playlistitems(format
):
1642 for string_segment
in format
.split(','):
1643 if '-' in string_segment
:
1644 start
, end
= string_segment
.split('-')
1645 for item
in range(int(start
), int(end
) + 1):
1648 yield int(string_segment
)
1649 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1651 ie_entries
= ie_result
['entries']
1652 if isinstance(ie_entries
, list):
1653 playlist_count
= len(ie_entries
)
1654 msg
= f
'Collected {playlist_count} videos; downloading %d of them'
1655 ie_result
['playlist_count'] = ie_result
.get('playlist_count') or playlist_count
1658 return ie_entries
[i
- 1]
1660 msg
= 'Downloading %d videos'
1661 if not isinstance(ie_entries
, (PagedList
, LazyList
)):
1662 ie_entries
= LazyList(ie_entries
)
1665 return YoutubeDL
.__handle
_extraction
_exceptions
(
1666 lambda self
, i
: ie_entries
[i
- 1]
1669 entries
, broken
= [], False
1670 items
= playlistitems
if playlistitems
is not None else itertools
.count(playliststart
)
1674 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1678 entry
= get_entry(i
)
1679 if entry
is MissingEntry
:
1680 raise EntryNotInPlaylist()
1681 except (IndexError, EntryNotInPlaylist
):
1682 if incomplete_entries
:
1683 raise EntryNotInPlaylist(f
'Entry {i} cannot be found')
1684 elif not playlistitems
:
1686 entries
.append(entry
)
1688 if entry
is not None:
1689 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1690 except (ExistingVideoReached
, RejectedVideoReached
):
1693 ie_result
['entries'] = entries
1695 # Save playlist_index before re-ordering
1697 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1698 for i
, entry
in enumerate(entries
, 1)
1699 if entry
is not None]
1700 n_entries
= len(entries
)
1702 if not (ie_result
.get('playlist_count') or broken
or playlistitems
or playlistend
):
1703 ie_result
['playlist_count'] = n_entries
1705 if not playlistitems
and (playliststart
!= 1 or playlistend
):
1706 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1707 ie_result
['requested_entries'] = playlistitems
1709 _infojson_written
= False
1710 if not self
.params
.get('simulate') and self
.params
.get('allow_playlist_files', True):
1711 ie_copy
= self
._playlist
_infodict
(ie_result
, n_entries
=n_entries
)
1712 _infojson_written
= self
._write
_info
_json
(
1713 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1714 if _infojson_written
is None:
1716 if self
._write
_description
('playlist', ie_result
,
1717 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1719 # TODO: This should be passed to ThumbnailsConvertor if necessary
1720 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1722 if self
.params
.get('playlistreverse', False):
1723 entries
= entries
[::-1]
1724 if self
.params
.get('playlistrandom', False):
1725 random
.shuffle(entries
)
1727 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1729 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1731 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1732 for i
, entry_tuple
in enumerate(entries
, 1):
1733 playlist_index
, entry
= entry_tuple
1734 if 'playlist-index' in self
.params
.get('compat_opts', []):
1735 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1
1736 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1737 # This __x_forwarded_for_ip thing is a bit ugly but requires
1740 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1742 'n_entries': n_entries
,
1743 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1744 'playlist_count': ie_result
.get('playlist_count'),
1745 'playlist_index': playlist_index
,
1746 'playlist_autonumber': i
,
1747 'playlist': playlist
,
1748 'playlist_id': ie_result
.get('id'),
1749 'playlist_title': ie_result
.get('title'),
1750 'playlist_uploader': ie_result
.get('uploader'),
1751 'playlist_uploader_id': ie_result
.get('uploader_id'),
1752 'extractor': ie_result
['extractor'],
1753 'webpage_url': ie_result
['webpage_url'],
1754 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1755 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1756 'extractor_key': ie_result
['extractor_key'],
1759 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1762 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1763 if not entry_result
:
1765 if failures
>= max_failures
:
1767 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1769 playlist_results
.append(entry_result
)
1770 ie_result
['entries'] = playlist_results
1772 # Write the updated info to json
1773 if _infojson_written
and self
._write
_info
_json
(
1774 'updated playlist', ie_result
,
1775 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1778 ie_result
= self
.run_all_pps('playlist', ie_result
)
1779 self
.to_screen(f
'[download] Finished downloading playlist: {playlist}')
1782 @__handle_extraction_exceptions
1783 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1784 return self
.process_ie_result(
1785 entry
, download
=download
, extra_info
=extra_info
)
1787 def _build_format_filter(self
, filter_spec
):
1788 " Returns a function to filter the formats according to the filter_spec "
1798 operator_rex
= re
.compile(r
'''(?x)\s*
1799 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1800 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1801 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1802 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1803 m
= operator_rex
.fullmatch(filter_spec
)
1806 comparison_value
= int(m
.group('value'))
1808 comparison_value
= parse_filesize(m
.group('value'))
1809 if comparison_value
is None:
1810 comparison_value
= parse_filesize(m
.group('value') + 'B')
1811 if comparison_value
is None:
1813 'Invalid value %r in format specification %r' % (
1814 m
.group('value'), filter_spec
))
1815 op
= OPERATORS
[m
.group('op')]
1820 '^=': lambda attr
, value
: attr
.startswith(value
),
1821 '$=': lambda attr
, value
: attr
.endswith(value
),
1822 '*=': lambda attr
, value
: value
in attr
,
1824 str_operator_rex
= re
.compile(r
'''(?x)\s*
1825 (?P<key>[a-zA-Z0-9._-]+)\s*
1826 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1827 (?P<value>[a-zA-Z0-9._-]+)\s*
1828 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1829 m
= str_operator_rex
.fullmatch(filter_spec
)
1831 comparison_value
= m
.group('value')
1832 str_op
= STR_OPERATORS
[m
.group('op')]
1833 if m
.group('negation'):
1834 op
= lambda attr
, value
: not str_op(attr
, value
)
1839 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1842 actual_value
= f
.get(m
.group('key'))
1843 if actual_value
is None:
1844 return m
.group('none_inclusive')
1845 return op(actual_value
, comparison_value
)
1848 def _check_formats(self
, formats
):
1850 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1851 path
= self
.get_output_path('temp')
1852 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1854 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1857 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1858 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1861 if os
.path
.exists(temp_file
.name
):
1863 os
.remove(temp_file
.name
)
1865 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1869 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1871 def _default_format_spec(self
, info_dict
, download
=True):
1874 merger
= FFmpegMergerPP(self
)
1875 return merger
.available
and merger
.can_merge()
1878 not self
.params
.get('simulate')
1882 or info_dict
.get('is_live', False)
1883 or self
.outtmpl_dict
['default'] == '-'))
1886 or self
.params
.get('allow_multiple_audio_streams', False)
1887 or 'format-spec' in self
.params
.get('compat_opts', []))
1890 'best/bestvideo+bestaudio' if prefer_best
1891 else 'bestvideo*+bestaudio/best' if not compat
1892 else 'bestvideo+bestaudio/best')
1894 def build_format_selector(self
, format_spec
):
1895 def syntax_error(note
, start
):
1897 'Invalid format specification: '
1898 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1899 return SyntaxError(message
)
1901 PICKFIRST
= 'PICKFIRST'
1905 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1907 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1908 'video': self
.params
.get('allow_multiple_video_streams', False)}
1910 check_formats
= self
.params
.get('check_formats') == 'selected'
1912 def _parse_filter(tokens
):
1914 for type, string
, start
, _
, _
in tokens
:
1915 if type == tokenize
.OP
and string
== ']':
1916 return ''.join(filter_parts
)
1918 filter_parts
.append(string
)
1920 def _remove_unused_ops(tokens
):
1921 # Remove operators that we don't use and join them with the surrounding strings
1922 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1923 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1924 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1925 for type, string
, start
, end
, line
in tokens
:
1926 if type == tokenize
.OP
and string
== '[':
1928 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1930 yield type, string
, start
, end
, line
1931 # everything inside brackets will be handled by _parse_filter
1932 for type, string
, start
, end
, line
in tokens
:
1933 yield type, string
, start
, end
, line
1934 if type == tokenize
.OP
and string
== ']':
1936 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1938 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1940 yield type, string
, start
, end
, line
1941 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1943 last_string
= string
1947 last_string
+= string
1949 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1951 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1953 current_selector
= None
1954 for type, string
, start
, _
, _
in tokens
:
1955 # ENCODING is only defined in python 3.x
1956 if type == getattr(tokenize
, 'ENCODING', None):
1958 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1959 current_selector
= FormatSelector(SINGLE
, string
, [])
1960 elif type == tokenize
.OP
:
1962 if not inside_group
:
1963 # ')' will be handled by the parentheses group
1964 tokens
.restore_last_token()
1966 elif inside_merge
and string
in ['/', ',']:
1967 tokens
.restore_last_token()
1969 elif inside_choice
and string
== ',':
1970 tokens
.restore_last_token()
1973 if not current_selector
:
1974 raise syntax_error('"," must follow a format selector', start
)
1975 selectors
.append(current_selector
)
1976 current_selector
= None
1978 if not current_selector
:
1979 raise syntax_error('"/" must follow a format selector', start
)
1980 first_choice
= current_selector
1981 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1982 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1984 if not current_selector
:
1985 current_selector
= FormatSelector(SINGLE
, 'best', [])
1986 format_filter
= _parse_filter(tokens
)
1987 current_selector
.filters
.append(format_filter
)
1989 if current_selector
:
1990 raise syntax_error('Unexpected "("', start
)
1991 group
= _parse_format_selection(tokens
, inside_group
=True)
1992 current_selector
= FormatSelector(GROUP
, group
, [])
1994 if not current_selector
:
1995 raise syntax_error('Unexpected "+"', start
)
1996 selector_1
= current_selector
1997 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1999 raise syntax_error('Expected a selector', start
)
2000 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2002 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
2003 elif type == tokenize
.ENDMARKER
:
2005 if current_selector
:
2006 selectors
.append(current_selector
)
2009 def _merge(formats_pair
):
2010 format_1
, format_2
= formats_pair
2013 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2014 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2016 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2017 get_no_more
= {'video': False, 'audio': False}
2018 for (i
, fmt_info
) in enumerate(formats_info
):
2019 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2022 for aud_vid
in ['audio', 'video']:
2023 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2024 if get_no_more
[aud_vid
]:
2027 get_no_more
[aud_vid
] = True
2029 if len(formats_info
) == 1:
2030 return formats_info
[0]
2032 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2033 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2035 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2036 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2038 output_ext
= self
.params
.get('merge_output_format')
2041 output_ext
= the_only_video
['ext']
2042 elif the_only_audio
and not video_fmts
:
2043 output_ext
= the_only_audio
['ext']
2047 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2050 'requested_formats': formats_info
,
2051 'format': '+'.join(filtered('format')),
2052 'format_id': '+'.join(filtered('format_id')),
2054 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2055 'language': '+'.join(orderedSet(filtered('language'))) or None,
2056 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2057 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2058 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2063 'width': the_only_video
.get('width'),
2064 'height': the_only_video
.get('height'),
2065 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2066 'fps': the_only_video
.get('fps'),
2067 'dynamic_range': the_only_video
.get('dynamic_range'),
2068 'vcodec': the_only_video
.get('vcodec'),
2069 'vbr': the_only_video
.get('vbr'),
2070 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2075 'acodec': the_only_audio
.get('acodec'),
2076 'abr': the_only_audio
.get('abr'),
2077 'asr': the_only_audio
.get('asr'),
2082 def _check_formats(formats
):
2083 if not check_formats
:
2086 yield from self
._check
_formats
(formats
)
2088 def _build_selector_function(selector
):
2089 if isinstance(selector
, list): # ,
2090 fs
= [_build_selector_function(s
) for s
in selector
]
2092 def selector_function(ctx
):
2095 return selector_function
2097 elif selector
.type == GROUP
: # ()
2098 selector_function
= _build_selector_function(selector
.selector
)
2100 elif selector
.type == PICKFIRST
: # /
2101 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2103 def selector_function(ctx
):
2105 picked_formats
= list(f(ctx
))
2107 return picked_formats
2110 elif selector
.type == MERGE
: # +
2111 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2113 def selector_function(ctx
):
2114 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2117 elif selector
.type == SINGLE
: # atom
2118 format_spec
= selector
.selector
or 'best'
2120 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2121 if format_spec
== 'all':
2122 def selector_function(ctx
):
2123 yield from _check_formats(ctx
['formats'][::-1])
2124 elif format_spec
== 'mergeall':
2125 def selector_function(ctx
):
2126 formats
= list(_check_formats(ctx
['formats']))
2129 merged_format
= formats
[-1]
2130 for f
in formats
[-2::-1]:
2131 merged_format
= _merge((merged_format
, f
))
2135 format_fallback
, format_reverse
, format_idx
= False, True, 1
2137 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2139 if mobj
is not None:
2140 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2141 format_reverse
= mobj
.group('bw')[0] == 'b'
2142 format_type
= (mobj
.group('type') or [None])[0]
2143 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2144 format_modified
= mobj
.group('mod') is not None
2146 format_fallback
= not format_type
and not format_modified
# for b, w
2148 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2149 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2150 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2151 if format_type
# bv, ba, wv, wa
2152 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2153 if not format_modified
# b, w
2154 else lambda f
: True) # b*, w*
2155 filter_f
= lambda f
: _filter_f(f
) and (
2156 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2158 if format_spec
in self
._format
_selection
_exts
['audio']:
2159 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2160 elif format_spec
in self
._format
_selection
_exts
['video']:
2161 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2162 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2163 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2165 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2167 def selector_function(ctx
):
2168 formats
= list(ctx
['formats'])
2169 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2170 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
2171 # for extractors with incomplete formats (audio only (soundcloud)
2172 # or video only (imgur)) best/worst will fallback to
2173 # best/worst {video,audio}-only format
2175 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2177 yield matches
[format_idx
- 1]
2181 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2183 def final_selector(ctx
):
2184 ctx_copy
= dict(ctx
)
2185 for _filter
in filters
:
2186 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2187 return selector_function(ctx_copy
)
2188 return final_selector
2190 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
2192 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
2193 except tokenize
.TokenError
:
2194 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2196 class TokenIterator(object):
2197 def __init__(self
, tokens
):
2198 self
.tokens
= tokens
2205 if self
.counter
>= len(self
.tokens
):
2206 raise StopIteration()
2207 value
= self
.tokens
[self
.counter
]
2213 def restore_last_token(self
):
2216 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2217 return _build_selector_function(parsed_selector
)
2219 def _calc_headers(self
, info_dict
):
2220 res
= std_headers
.copy()
2222 add_headers
= info_dict
.get('http_headers')
2224 res
.update(add_headers
)
2226 cookies
= self
._calc
_cookies
(info_dict
)
2228 res
['Cookie'] = cookies
2230 if 'X-Forwarded-For' not in res
:
2231 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2232 if x_forwarded_for_ip
:
2233 res
['X-Forwarded-For'] = x_forwarded_for_ip
2237 def _calc_cookies(self
, info_dict
):
2238 pr
= sanitized_Request(info_dict
['url'])
2239 self
.cookiejar
.add_cookie_header(pr
)
2240 return pr
.get_header('Cookie')
2242 def _sort_thumbnails(self
, thumbnails
):
2243 thumbnails
.sort(key
=lambda t
: (
2244 t
.get('preference') if t
.get('preference') is not None else -1,
2245 t
.get('width') if t
.get('width') is not None else -1,
2246 t
.get('height') if t
.get('height') is not None else -1,
2247 t
.get('id') if t
.get('id') is not None else '',
2250 def _sanitize_thumbnails(self
, info_dict
):
2251 thumbnails
= info_dict
.get('thumbnails')
2252 if thumbnails
is None:
2253 thumbnail
= info_dict
.get('thumbnail')
2255 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2259 def check_thumbnails(thumbnails
):
2260 for t
in thumbnails
:
2261 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2263 self
.urlopen(HEADRequest(t
['url']))
2264 except network_exceptions
as err
:
2265 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2269 self
._sort
_thumbnails
(thumbnails
)
2270 for i
, t
in enumerate(thumbnails
):
2271 if t
.get('id') is None:
2273 if t
.get('width') and t
.get('height'):
2274 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2275 t
['url'] = sanitize_url(t
['url'])
2277 if self
.params
.get('check_formats') is True:
2278 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2280 info_dict
['thumbnails'] = thumbnails
2282 def process_video_result(self
, info_dict
, download
=True):
2283 assert info_dict
.get('_type', 'video') == 'video'
2284 self
._num
_videos
+= 1
2286 if 'id' not in info_dict
:
2287 raise ExtractorError('Missing "id" field in extractor result')
2288 if 'title' not in info_dict
:
2289 raise ExtractorError('Missing "title" field in extractor result',
2290 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2292 def report_force_conversion(field
, field_not
, conversion
):
2293 self
.report_warning(
2294 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2295 % (field
, field_not
, conversion
))
2297 def sanitize_string_field(info
, string_field
):
2298 field
= info
.get(string_field
)
2299 if field
is None or isinstance(field
, compat_str
):
2301 report_force_conversion(string_field
, 'a string', 'string')
2302 info
[string_field
] = compat_str(field
)
2304 def sanitize_numeric_fields(info
):
2305 for numeric_field
in self
._NUMERIC
_FIELDS
:
2306 field
= info
.get(numeric_field
)
2307 if field
is None or isinstance(field
, compat_numeric_types
):
2309 report_force_conversion(numeric_field
, 'numeric', 'int')
2310 info
[numeric_field
] = int_or_none(field
)
2312 sanitize_string_field(info_dict
, 'id')
2313 sanitize_numeric_fields(info_dict
)
2315 if 'playlist' not in info_dict
:
2316 # It isn't part of a playlist
2317 info_dict
['playlist'] = None
2318 info_dict
['playlist_index'] = None
2320 self
._sanitize
_thumbnails
(info_dict
)
2322 thumbnail
= info_dict
.get('thumbnail')
2323 thumbnails
= info_dict
.get('thumbnails')
2325 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2327 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2329 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2330 info_dict
['display_id'] = info_dict
['id']
2332 if info_dict
.get('duration') is not None:
2333 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2335 for ts_key
, date_key
in (
2336 ('timestamp', 'upload_date'),
2337 ('release_timestamp', 'release_date'),
2338 ('modified_timestamp', 'modified_date'),
2340 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2341 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2342 # see http://bugs.python.org/issue1646728)
2344 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2345 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2346 except (ValueError, OverflowError, OSError):
2349 live_keys
= ('is_live', 'was_live')
2350 live_status
= info_dict
.get('live_status')
2351 if live_status
is None:
2352 for key
in live_keys
:
2353 if info_dict
.get(key
) is False:
2355 if info_dict
.get(key
):
2358 if all(info_dict
.get(key
) is False for key
in live_keys
):
2359 live_status
= 'not_live'
2361 info_dict
['live_status'] = live_status
2362 for key
in live_keys
:
2363 if info_dict
.get(key
) is None:
2364 info_dict
[key
] = (live_status
== key
)
2366 # Auto generate title fields corresponding to the *_number fields when missing
2367 # in order to always have clean titles. This is very common for TV series.
2368 for field
in ('chapter', 'season', 'episode'):
2369 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2370 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2372 for cc_kind
in ('subtitles', 'automatic_captions'):
2373 cc
= info_dict
.get(cc_kind
)
2375 for _
, subtitle
in cc
.items():
2376 for subtitle_format
in subtitle
:
2377 if subtitle_format
.get('url'):
2378 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2379 if subtitle_format
.get('ext') is None:
2380 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2382 automatic_captions
= info_dict
.get('automatic_captions')
2383 subtitles
= info_dict
.get('subtitles')
2385 info_dict
['requested_subtitles'] = self
.process_subtitles(
2386 info_dict
['id'], subtitles
, automatic_captions
)
2388 if info_dict
.get('formats') is None:
2389 # There's only one format available
2390 formats
= [info_dict
]
2392 formats
= info_dict
['formats']
2394 info_dict
['__has_drm'] = any(f
.get('has_drm') for f
in formats
)
2395 if not self
.params
.get('allow_unplayable_formats'):
2396 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2398 # backward compatibility
2399 info_dict
['fulltitle'] = info_dict
['title']
2401 if info_dict
.get('is_live'):
2402 get_from_start
= bool(self
.params
.get('live_from_start'))
2403 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2404 if not get_from_start
:
2405 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2408 self
.raise_no_formats(info_dict
)
2410 def is_wellformed(f
):
2413 self
.report_warning(
2414 '"url" field is missing or empty - skipping format, '
2415 'there is an error in extractor')
2417 if isinstance(url
, bytes):
2418 sanitize_string_field(f
, 'url')
2421 # Filter out malformed formats for better extraction robustness
2422 formats
= list(filter(is_wellformed
, formats
))
2426 # We check that all the formats have the format and format_id fields
2427 for i
, format
in enumerate(formats
):
2428 sanitize_string_field(format
, 'format_id')
2429 sanitize_numeric_fields(format
)
2430 format
['url'] = sanitize_url(format
['url'])
2431 if not format
.get('format_id'):
2432 format
['format_id'] = compat_str(i
)
2434 # Sanitize format_id from characters used in format selector expression
2435 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2436 format_id
= format
['format_id']
2437 if format_id
not in formats_dict
:
2438 formats_dict
[format_id
] = []
2439 formats_dict
[format_id
].append(format
)
2441 # Make sure all formats have unique format_id
2442 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2443 for format_id
, ambiguous_formats
in formats_dict
.items():
2444 ambigious_id
= len(ambiguous_formats
) > 1
2445 for i
, format
in enumerate(ambiguous_formats
):
2447 format
['format_id'] = '%s-%d' % (format_id
, i
)
2448 if format
.get('ext') is None:
2449 format
['ext'] = determine_ext(format
['url']).lower()
2450 # Ensure there is no conflict between id and ext in format selection
2451 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2452 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2453 format
['format_id'] = 'f%s' % format
['format_id']
2455 for i
, format
in enumerate(formats
):
2456 if format
.get('format') is None:
2457 format
['format'] = '{id} - {res}{note}'.format(
2458 id=format
['format_id'],
2459 res
=self
.format_resolution(format
),
2460 note
=format_field(format
, 'format_note', ' (%s)'),
2462 if format
.get('protocol') is None:
2463 format
['protocol'] = determine_protocol(format
)
2464 if format
.get('resolution') is None:
2465 format
['resolution'] = self
.format_resolution(format
, default
=None)
2466 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2467 format
['dynamic_range'] = 'SDR'
2468 if (info_dict
.get('duration') and format
.get('tbr')
2469 and not format
.get('filesize') and not format
.get('filesize_approx')):
2470 format
['filesize_approx'] = info_dict
['duration'] * format
['tbr'] * (1024 / 8)
2472 # Add HTTP headers, so that external programs can use them from the
2474 full_format_info
= info_dict
.copy()
2475 full_format_info
.update(format
)
2476 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2477 # Remove private housekeeping stuff
2478 if '__x_forwarded_for_ip' in info_dict
:
2479 del info_dict
['__x_forwarded_for_ip']
2481 # TODO Central sorting goes here
2483 if self
.params
.get('check_formats') is True:
2484 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2486 if not formats
or formats
[0] is not info_dict
:
2487 # only set the 'formats' fields if the original info_dict list them
2488 # otherwise we end up with a circular reference, the first (and unique)
2489 # element in the 'formats' field in info_dict is info_dict itself,
2490 # which can't be exported to json
2491 info_dict
['formats'] = formats
2493 info_dict
, _
= self
.pre_process(info_dict
)
2495 # The pre-processors may have modified the formats
2496 formats
= info_dict
.get('formats', [info_dict
])
2498 list_only
= self
.params
.get('simulate') is None and (
2499 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2500 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2501 if self
.params
.get('list_thumbnails'):
2502 self
.list_thumbnails(info_dict
)
2503 if self
.params
.get('listsubtitles'):
2504 if 'automatic_captions' in info_dict
:
2505 self
.list_subtitles(
2506 info_dict
['id'], automatic_captions
, 'automatic captions')
2507 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2508 if self
.params
.get('listformats') or interactive_format_selection
:
2509 self
.list_formats(info_dict
)
2511 # Without this printing, -F --print-json will not work
2512 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2515 format_selector
= self
.format_selector
2516 if format_selector
is None:
2517 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2518 self
.write_debug('Default format spec: %s' % req_format
)
2519 format_selector
= self
.build_format_selector(req_format
)
2522 if interactive_format_selection
:
2524 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2526 format_selector
= self
.build_format_selector(req_format
)
2527 except SyntaxError as err
:
2528 self
.report_error(err
, tb
=False, is_error
=False)
2531 # While in format selection we may need to have an access to the original
2532 # format set in order to calculate some metrics or do some processing.
2533 # For now we need to be able to guess whether original formats provided
2534 # by extractor are incomplete or not (i.e. whether extractor provides only
2535 # video-only or audio-only formats) for proper formats selection for
2536 # extractors with such incomplete formats (see
2537 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2538 # Since formats may be filtered during format selection and may not match
2539 # the original formats the results may be incorrect. Thus original formats
2540 # or pre-calculated metrics should be passed to format selection routines
2542 # We will pass a context object containing all necessary additional data
2543 # instead of just formats.
2544 # This fixes incorrect format selection issue (see
2545 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2546 incomplete_formats
= (
2547 # All formats are video-only or
2548 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2549 # all formats are audio-only
2550 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2554 'incomplete_formats': incomplete_formats
,
2557 formats_to_download
= list(format_selector(ctx
))
2558 if interactive_format_selection
and not formats_to_download
:
2559 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2563 if not formats_to_download
:
2564 if not self
.params
.get('ignore_no_formats_error'):
2565 raise ExtractorError('Requested format is not available', expected
=True,
2566 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2567 self
.report_warning('Requested format is not available')
2568 # Process what we can, even without any available formats.
2569 formats_to_download
= [{}]
2571 best_format
= formats_to_download
[-1]
2575 f
'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2576 + ', '.join([f
['format_id'] for f
in formats_to_download
]))
2577 max_downloads_reached
= False
2578 for i
, fmt
in enumerate(formats_to_download
):
2579 formats_to_download
[i
] = new_info
= dict(info_dict
)
2580 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2581 new_info
.update(fmt
)
2582 new_info
['__original_infodict'] = info_dict
2584 self
.process_info(new_info
)
2585 except MaxDownloadsReached
:
2586 max_downloads_reached
= True
2587 new_info
.pop('__original_infodict')
2588 # Remove copied info
2589 for key
, val
in tuple(new_info
.items()):
2590 if info_dict
.get(key
) == val
:
2592 if max_downloads_reached
:
2595 write_archive
= set(f
.get('__write_download_archive', False) for f
in formats_to_download
)
2596 assert write_archive
.issubset({True, False, 'ignore'}
)
2597 if True in write_archive
and False not in write_archive
:
2598 self
.record_download_archive(info_dict
)
2600 info_dict
['requested_downloads'] = formats_to_download
2601 info_dict
= self
.run_all_pps('after_video', info_dict
)
2602 if max_downloads_reached
:
2603 raise MaxDownloadsReached()
2605 # We update the info dict with the selected best quality format (backwards compatibility)
2606 info_dict
.update(best_format
)
2609 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2610 """Select the requested subtitles and their format"""
2612 if normal_subtitles
and self
.params
.get('writesubtitles'):
2613 available_subs
.update(normal_subtitles
)
2614 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2615 for lang
, cap_info
in automatic_captions
.items():
2616 if lang
not in available_subs
:
2617 available_subs
[lang
] = cap_info
2619 if (not self
.params
.get('writesubtitles') and not
2620 self
.params
.get('writeautomaticsub') or not
2624 all_sub_langs
= available_subs
.keys()
2625 if self
.params
.get('allsubtitles', False):
2626 requested_langs
= all_sub_langs
2627 elif self
.params
.get('subtitleslangs', False):
2628 # A list is used so that the order of languages will be the same as
2629 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2630 requested_langs
= []
2631 for lang_re
in self
.params
.get('subtitleslangs'):
2632 if lang_re
== 'all':
2633 requested_langs
.extend(all_sub_langs
)
2635 discard
= lang_re
[0] == '-'
2637 lang_re
= lang_re
[1:]
2638 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2640 for lang
in current_langs
:
2641 while lang
in requested_langs
:
2642 requested_langs
.remove(lang
)
2644 requested_langs
.extend(current_langs
)
2645 requested_langs
= orderedSet(requested_langs
)
2646 elif 'en' in available_subs
:
2647 requested_langs
= ['en']
2649 requested_langs
= [list(all_sub_langs
)[0]]
2651 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2653 formats_query
= self
.params
.get('subtitlesformat', 'best')
2654 formats_preference
= formats_query
.split('/') if formats_query
else []
2656 for lang
in requested_langs
:
2657 formats
= available_subs
.get(lang
)
2659 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2661 for ext
in formats_preference
:
2665 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2671 self
.report_warning(
2672 'No subtitle format found matching "%s" for language %s, '
2673 'using %s' % (formats_query
, lang
, f
['ext']))
2677 def _forceprint(self
, tmpl
, info_dict
):
2678 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2679 if mobj
and mobj
.group(1):
2680 tmpl
= f
'{tmpl[:-1]} = %({tmpl[:-1]})s'
2682 tmpl
= '%({})s'.format(tmpl
)
2684 info_dict
= info_dict
.copy()
2685 info_dict
['formats_table'] = self
.render_formats_table(info_dict
)
2686 info_dict
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2687 info_dict
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2688 info_dict
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2689 self
.to_stdout(self
.evaluate_outtmpl(tmpl
, info_dict
))
2691 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2692 def print_mandatory(field
, actual_field
=None):
2693 if actual_field
is None:
2694 actual_field
= field
2695 if (self
.params
.get('force%s' % field
, False)
2696 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2697 self
.to_stdout(info_dict
[actual_field
])
2699 def print_optional(field
):
2700 if (self
.params
.get('force%s' % field
, False)
2701 and info_dict
.get(field
) is not None):
2702 self
.to_stdout(info_dict
[field
])
2704 info_dict
= info_dict
.copy()
2705 if filename
is not None:
2706 info_dict
['filename'] = filename
2707 if info_dict
.get('requested_formats') is not None:
2708 # For RTMP URLs, also include the playpath
2709 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2710 elif 'url' in info_dict
:
2711 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2713 if self
.params
['forceprint'].get('video') or self
.params
.get('forcejson'):
2714 self
.post_extract(info_dict
)
2715 for tmpl
in self
.params
['forceprint'].get('video', []):
2716 self
._forceprint
(tmpl
, info_dict
)
2718 print_mandatory('title')
2719 print_mandatory('id')
2720 print_mandatory('url', 'urls')
2721 print_optional('thumbnail')
2722 print_optional('description')
2723 print_optional('filename')
2724 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2725 self
.to_stdout(formatSeconds(info_dict
['duration']))
2726 print_mandatory('format')
2728 if self
.params
.get('forcejson'):
2729 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2731 def dl(self
, name
, info
, subtitle
=False, test
=False):
2732 if not info
.get('url'):
2733 self
.raise_no_formats(info
, True)
2736 verbose
= self
.params
.get('verbose')
2739 'quiet': self
.params
.get('quiet') or not verbose
,
2741 'noprogress': not verbose
,
2743 'skip_unavailable_fragments': False,
2744 'keep_fragments': False,
2746 '_no_ytdl_file': True,
2749 params
= self
.params
2750 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2752 for ph
in self
._progress
_hooks
:
2753 fd
.add_progress_hook(ph
)
2754 urls
= '", "'.join([f
['url'] for f
in info
.get('requested_formats', [])] or [info
['url']])
2755 self
.write_debug('Invoking downloader on "%s"' % urls
)
2757 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2758 # But it may contain objects that are not deep-copyable
2759 new_info
= self
._copy
_infodict
(info
)
2760 if new_info
.get('http_headers') is None:
2761 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2762 return fd
.download(name
, new_info
, subtitle
)
2764 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2765 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2766 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2767 return existing_files
[0]
2769 for file in existing_files
:
2770 self
.report_file_delete(file)
2774 def process_info(self
, info_dict
):
2775 """Process a single resolved IE result. (Modified it in-place)"""
2777 assert info_dict
.get('_type', 'video') == 'video'
2778 original_infodict
= info_dict
2780 if 'format' not in info_dict
and 'ext' in info_dict
:
2781 info_dict
['format'] = info_dict
['ext']
2783 if self
._match
_entry
(info_dict
) is not None:
2784 info_dict
['__write_download_archive'] = 'ignore'
2787 self
.post_extract(info_dict
)
2788 self
._num
_downloads
+= 1
2790 # info_dict['_filename'] needs to be set for backward compatibility
2791 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2792 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2796 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2798 if self
.params
.get('simulate'):
2799 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2802 if full_filename
is None:
2804 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2806 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2809 if self
._write
_description
('video', info_dict
,
2810 self
.prepare_filename(info_dict
, 'description')) is None:
2813 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2814 if sub_files
is None:
2816 files_to_move
.update(dict(sub_files
))
2818 thumb_files
= self
._write
_thumbnails
(
2819 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2820 if thumb_files
is None:
2822 files_to_move
.update(dict(thumb_files
))
2824 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2825 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2826 if _infojson_written
:
2827 info_dict
['infojson_filename'] = infofn
2828 # For backward compatibility, even though it was a private field
2829 info_dict
['__infojson_filename'] = infofn
2830 elif _infojson_written
is None:
2833 # Note: Annotations are deprecated
2835 if self
.params
.get('writeannotations', False):
2836 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2838 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2840 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2841 self
.to_screen('[info] Video annotations are already present')
2842 elif not info_dict
.get('annotations'):
2843 self
.report_warning('There are no annotations to write.')
2846 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2847 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2848 annofile
.write(info_dict
['annotations'])
2849 except (KeyError, TypeError):
2850 self
.report_warning('There are no annotations to write.')
2851 except (OSError, IOError):
2852 self
.report_error('Cannot write annotations file: ' + annofn
)
2855 # Write internet shortcut files
2856 def _write_link_file(link_type
):
2857 if 'webpage_url' not in info_dict
:
2858 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2860 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2861 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
2863 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2864 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2867 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2868 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2869 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2870 template_vars
= {'url': iri_to_uri(info_dict['webpage_url'])}
2871 if link_type
== 'desktop':
2872 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2873 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
2874 except (OSError, IOError):
2875 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
2880 'url': self
.params
.get('writeurllink'),
2881 'webloc': self
.params
.get('writewebloclink'),
2882 'desktop': self
.params
.get('writedesktoplink'),
2884 if self
.params
.get('writelink'):
2885 link_type
= ('webloc' if sys
.platform
== 'darwin'
2886 else 'desktop' if sys
.platform
.startswith('linux')
2888 write_links
[link_type
] = True
2890 if any(should_write
and not _write_link_file(link_type
)
2891 for link_type
, should_write
in write_links
.items()):
2894 def replace_info_dict(new_info
):
2896 if new_info
== info_dict
:
2899 info_dict
.update(new_info
)
2902 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2903 replace_info_dict(new_info
)
2904 except PostProcessingError
as err
:
2905 self
.report_error('Preprocessing: %s' % str(err
))
2908 if self
.params
.get('skip_download'):
2909 info_dict
['filepath'] = temp_filename
2910 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2911 info_dict
['__files_to_move'] = files_to_move
2912 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
2913 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2916 info_dict
.setdefault('__postprocessors', [])
2919 def existing_video_file(*filepaths
):
2920 ext
= info_dict
.get('ext')
2921 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
2922 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
2923 default_overwrite
=False)
2925 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
2929 if info_dict
.get('requested_formats') is not None:
2931 def compatible_formats(formats
):
2932 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2933 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2934 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2935 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2939 exts
= set(format
.get('ext') for format
in formats
)
2941 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2944 for ext_sets
in COMPATIBLE_EXTS
:
2945 if ext_sets
.issuperset(exts
):
2947 # TODO: Check acodec/vcodec
2950 requested_formats
= info_dict
['requested_formats']
2951 old_ext
= info_dict
['ext']
2952 if self
.params
.get('merge_output_format') is None:
2953 if not compatible_formats(requested_formats
):
2954 info_dict
['ext'] = 'mkv'
2955 self
.report_warning(
2956 'Requested formats are incompatible for merge and will be merged into mkv')
2957 if (info_dict
['ext'] == 'webm'
2958 and info_dict
.get('thumbnails')
2959 # check with type instead of pp_key, __name__, or isinstance
2960 # since we dont want any custom PPs to trigger this
2961 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])):
2962 info_dict
['ext'] = 'mkv'
2963 self
.report_warning(
2964 'webm doesn\'t support embedding a thumbnail, mkv will be used')
2965 new_ext
= info_dict
['ext']
2967 def correct_ext(filename
, ext
=new_ext
):
2970 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2972 os
.path
.splitext(filename
)[0]
2973 if filename_real_ext
in (old_ext
, new_ext
)
2975 return '%s.%s' % (filename_wo_ext
, ext
)
2977 # Ensure filename always has a correct extension for successful merge
2978 full_filename
= correct_ext(full_filename
)
2979 temp_filename
= correct_ext(temp_filename
)
2980 dl_filename
= existing_video_file(full_filename
, temp_filename
)
2981 info_dict
['__real_download'] = False
2984 merger
= FFmpegMergerPP(self
)
2986 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
2987 if dl_filename
is not None:
2988 self
.report_file_already_downloaded(dl_filename
)
2990 for f
in requested_formats
if fd
!= FFmpegFD
else []:
2991 f
['filepath'] = fname
= prepend_extension(
2992 correct_ext(temp_filename
, info_dict
['ext']),
2993 'f%s' % f
['format_id'], info_dict
['ext'])
2994 downloaded
.append(fname
)
2995 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
2996 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2997 info_dict
['__real_download'] = real_download
2999 if self
.params
.get('allow_unplayable_formats'):
3000 self
.report_warning(
3001 'You have requested merging of multiple formats '
3002 'while also allowing unplayable formats to be downloaded. '
3003 'The formats won\'t be merged to prevent data corruption.')
3004 elif not merger
.available
:
3005 self
.report_warning(
3006 'You have requested merging of multiple formats but ffmpeg is not installed. '
3007 'The formats won\'t be merged.')
3009 if temp_filename
== '-':
3010 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3011 else 'but the formats are incompatible for simultaneous download' if merger
.available
3012 else 'but ffmpeg is not installed')
3013 self
.report_warning(
3014 f
'You have requested downloading multiple formats to stdout {reason}. '
3015 'The formats will be streamed one after the other')
3016 fname
= temp_filename
3017 for f
in requested_formats
:
3018 new_info
= dict(info_dict
)
3019 del new_info
['requested_formats']
3021 if temp_filename
!= '-':
3022 fname
= prepend_extension(
3023 correct_ext(temp_filename
, new_info
['ext']),
3024 'f%s' % f
['format_id'], new_info
['ext'])
3025 if not self
._ensure
_dir
_exists
(fname
):
3027 f
['filepath'] = fname
3028 downloaded
.append(fname
)
3029 partial_success
, real_download
= self
.dl(fname
, new_info
)
3030 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3031 success
= success
and partial_success
3033 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3034 info_dict
['__postprocessors'].append(merger
)
3035 info_dict
['__files_to_merge'] = downloaded
3036 # Even if there were no downloads, it is being merged only now
3037 info_dict
['__real_download'] = True
3039 for file in downloaded
:
3040 files_to_move
[file] = None
3042 # Just a single file
3043 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3044 if dl_filename
is None or dl_filename
== temp_filename
:
3045 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3046 # So we should try to resume the download
3047 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3048 info_dict
['__real_download'] = real_download
3050 self
.report_file_already_downloaded(dl_filename
)
3052 dl_filename
= dl_filename
or temp_filename
3053 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3055 except network_exceptions
as err
:
3056 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3058 except (OSError, IOError) as err
:
3059 raise UnavailableVideoError(err
)
3060 except (ContentTooShortError
, ) as err
:
3061 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
3064 if success
and full_filename
!= '-':
3068 fixup_policy
= self
.params
.get('fixup')
3069 vid
= info_dict
['id']
3071 if fixup_policy
in ('ignore', 'never'):
3073 elif fixup_policy
== 'warn':
3075 elif fixup_policy
!= 'force':
3076 assert fixup_policy
in ('detect_or_warn', None)
3077 if not info_dict
.get('__real_download'):
3080 def ffmpeg_fixup(cndn
, msg
, cls
):
3084 self
.report_warning(f
'{vid}: {msg}')
3088 info_dict
['__postprocessors'].append(pp
)
3090 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3092 stretched_ratio
= info_dict
.get('stretched_ratio')
3094 stretched_ratio
not in (1, None),
3095 f
'Non-uniform pixel ratio {stretched_ratio}',
3096 FFmpegFixupStretchedPP
)
3099 (info_dict
.get('requested_formats') is None
3100 and info_dict
.get('container') == 'm4a_dash'
3101 and info_dict
.get('ext') == 'm4a'),
3102 'writing DASH m4a. Only some players support this container',
3105 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3106 downloader
= downloader
.__name
__ if downloader
else None
3108 if info_dict
.get('requested_formats') is None: # Not necessary if doing merger
3109 ffmpeg_fixup(downloader
== 'HlsFD',
3110 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3112 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3113 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3115 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3116 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP
)
3120 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3121 except PostProcessingError
as err
:
3122 self
.report_error('Postprocessing: %s' % str(err
))
3125 for ph
in self
._post
_hooks
:
3126 ph(info_dict
['filepath'])
3127 except Exception as err
:
3128 self
.report_error('post hooks: %s' % str(err
))
3130 info_dict
['__write_download_archive'] = True
3132 if self
.params
.get('force_write_download_archive'):
3133 info_dict
['__write_download_archive'] = True
3135 # Make sure the info_dict was modified in-place
3136 assert info_dict
is original_infodict
3138 max_downloads
= self
.params
.get('max_downloads')
3139 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
3140 raise MaxDownloadsReached()
3142 def __download_wrapper(self
, func
):
3143 @functools.wraps(func
)
3144 def wrapper(*args
, **kwargs
):
3146 res
= func(*args
, **kwargs
)
3147 except UnavailableVideoError
as e
:
3148 self
.report_error(e
)
3149 except MaxDownloadsReached
as e
:
3150 self
.to_screen(f
'[info] {e}')
3152 except DownloadCancelled
as e
:
3153 self
.to_screen(f
'[info] {e}')
3154 if not self
.params
.get('break_per_url'):
3157 if self
.params
.get('dump_single_json', False):
3158 self
.post_extract(res
)
3159 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3162 def download(self
, url_list
):
3163 """Download a given list of URLs."""
3164 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3165 outtmpl
= self
.outtmpl_dict
['default']
3166 if (len(url_list
) > 1
3168 and '%' not in outtmpl
3169 and self
.params
.get('max_downloads') != 1):
3170 raise SameFileError(outtmpl
)
3172 for url
in url_list
:
3173 self
.__download
_wrapper
(self
.extract_info
)(
3174 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3176 return self
._download
_retcode
3178 def download_with_info_file(self
, info_filename
):
3179 with contextlib
.closing(fileinput
.FileInput(
3180 [info_filename
], mode
='r',
3181 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3182 # FileInput doesn't have a read method, we can't call json.load
3183 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3185 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3186 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3187 if not isinstance(e
, EntryNotInPlaylist
):
3188 self
.to_stderr('\r')
3189 webpage_url
= info
.get('webpage_url')
3190 if webpage_url
is not None:
3191 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3192 return self
.download([webpage_url
])
3195 return self
._download
_retcode
3198 def sanitize_info(info_dict
, remove_private_keys
=False):
3199 ''' Sanitize the infodict for converting to json '''
3200 if info_dict
is None:
3202 info_dict
.setdefault('epoch', int(time
.time()))
3203 remove_keys
= {'__original_infodict'}
# Always remove this since this may contain a copy of the entire dict
3204 keep_keys
= ['_type'] # Always keep this to facilitate load-info-json
3205 if remove_private_keys
:
3207 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3208 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3210 reject
= lambda k
, v
: k
not in keep_keys
and (
3211 k
.startswith('_') or k
in remove_keys
or v
is None)
3213 reject
= lambda k
, v
: k
in remove_keys
3216 if isinstance(obj
, dict):
3217 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3218 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3219 return list(map(filter_fn
, obj
))
3220 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3225 return filter_fn(info_dict
)
3228 def filter_requested_info(info_dict
, actually_filter
=True):
3229 ''' Alias of sanitize_info for backward compatibility '''
3230 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3233 def post_extract(info_dict
):
3234 def actual_post_extract(info_dict
):
3235 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3236 for video_dict
in info_dict
.get('entries', {}):
3237 actual_post_extract(video_dict
or {})
3240 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
3241 extra
= post_extractor().items()
3242 info_dict
.update(extra
)
3243 info_dict
.pop('__post_extractor', None)
3245 original_infodict
= info_dict
.get('__original_infodict') or {}
3246 original_infodict
.update(extra
)
3247 original_infodict
.pop('__post_extractor', None)
3249 actual_post_extract(info_dict
or {})
3251 def run_pp(self
, pp
, infodict
):
3252 files_to_delete
= []
3253 if '__files_to_move' not in infodict
:
3254 infodict
['__files_to_move'] = {}
3256 files_to_delete
, infodict
= pp
.run(infodict
)
3257 except PostProcessingError
as e
:
3258 # Must be True and not 'only_download'
3259 if self
.params
.get('ignoreerrors') is True:
3260 self
.report_error(e
)
3264 if not files_to_delete
:
3266 if self
.params
.get('keepvideo', False):
3267 for f
in files_to_delete
:
3268 infodict
['__files_to_move'].setdefault(f
, '')
3270 for old_filename
in set(files_to_delete
):
3271 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
3273 os
.remove(encodeFilename(old_filename
))
3274 except (IOError, OSError):
3275 self
.report_warning('Unable to remove downloaded original file')
3276 if old_filename
in infodict
['__files_to_move']:
3277 del infodict
['__files_to_move'][old_filename
]
3280 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3281 for tmpl
in self
.params
['forceprint'].get(key
, []):
3282 self
._forceprint
(tmpl
, info
)
3283 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3284 info
= self
.run_pp(pp
, info
)
3287 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3288 info
= dict(ie_info
)
3289 info
['__files_to_move'] = files_to_move
or {}
3290 info
= self
.run_all_pps(key
, info
)
3291 return info
, info
.pop('__files_to_move', None)
3293 def post_process(self
, filename
, info
, files_to_move
=None):
3294 """Run all the postprocessors on the given file."""
3295 info
['filepath'] = filename
3296 info
['__files_to_move'] = files_to_move
or {}
3297 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3298 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3299 del info
['__files_to_move']
3300 return self
.run_all_pps('after_move', info
)
3302 def _make_archive_id(self
, info_dict
):
3303 video_id
= info_dict
.get('id')
3306 # Future-proof against any change in case
3307 # and backwards compatibility with prior versions
3308 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3309 if extractor
is None:
3310 url
= str_or_none(info_dict
.get('url'))
3313 # Try to find matching extractor for the URL and take its ie_key
3314 for ie_key
, ie
in self
._ies
.items():
3315 if ie
.suitable(url
):
3320 return '%s %s' % (extractor
.lower(), video_id
)
3322 def in_download_archive(self
, info_dict
):
3323 fn
= self
.params
.get('download_archive')
3327 vid_id
= self
._make
_archive
_id
(info_dict
)
3329 return False # Incomplete video information
3331 return vid_id
in self
.archive
3333 def record_download_archive(self
, info_dict
):
3334 fn
= self
.params
.get('download_archive')
3337 vid_id
= self
._make
_archive
_id
(info_dict
)
3339 self
.write_debug(f
'Adding to archive: {vid_id}')
3340 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3341 archive_file
.write(vid_id
+ '\n')
3342 self
.archive
.add(vid_id
)
3345 def format_resolution(format
, default
='unknown'):
3346 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3348 if format
.get('resolution') is not None:
3349 return format
['resolution']
3350 if format
.get('width') and format
.get('height'):
3351 return '%dx%d' % (format
['width'], format
['height'])
3352 elif format
.get('height'):
3353 return '%sp' % format
['height']
3354 elif format
.get('width'):
3355 return '%dx?' % format
['width']
3358 def _list_format_headers(self
, *headers
):
3359 if self
.params
.get('listformats_table', True) is not False:
3360 return [self
._format
_screen
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3363 def _format_note(self
, fdict
):
3365 if fdict
.get('ext') in ['f4f', 'f4m']:
3366 res
+= '(unsupported)'
3367 if fdict
.get('language'):
3370 res
+= '[%s]' % fdict
['language']
3371 if fdict
.get('format_note') is not None:
3374 res
+= fdict
['format_note']
3375 if fdict
.get('tbr') is not None:
3378 res
+= '%4dk' % fdict
['tbr']
3379 if fdict
.get('container') is not None:
3382 res
+= '%s container' % fdict
['container']
3383 if (fdict
.get('vcodec') is not None
3384 and fdict
.get('vcodec') != 'none'):
3387 res
+= fdict
['vcodec']
3388 if fdict
.get('vbr') is not None:
3390 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3392 if fdict
.get('vbr') is not None:
3393 res
+= '%4dk' % fdict
['vbr']
3394 if fdict
.get('fps') is not None:
3397 res
+= '%sfps' % fdict
['fps']
3398 if fdict
.get('acodec') is not None:
3401 if fdict
['acodec'] == 'none':
3404 res
+= '%-5s' % fdict
['acodec']
3405 elif fdict
.get('abr') is not None:
3409 if fdict
.get('abr') is not None:
3410 res
+= '@%3dk' % fdict
['abr']
3411 if fdict
.get('asr') is not None:
3412 res
+= ' (%5dHz)' % fdict
['asr']
3413 if fdict
.get('filesize') is not None:
3416 res
+= format_bytes(fdict
['filesize'])
3417 elif fdict
.get('filesize_approx') is not None:
3420 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3423 def render_formats_table(self
, info_dict
):
3424 if not info_dict
.get('formats') and not info_dict
.get('url'):
3427 formats
= info_dict
.get('formats', [info_dict
])
3428 if not self
.params
.get('listformats_table', True) is not False:
3431 format_field(f
, 'format_id'),
3432 format_field(f
, 'ext'),
3433 self
.format_resolution(f
),
3434 self
._format
_note
(f
)
3435 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3436 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3438 delim
= self
._format
_screen
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3441 self
._format
_screen
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3442 format_field(f
, 'ext'),
3443 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3444 format_field(f
, 'fps', '\t%d'),
3445 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3447 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3448 format_field(f
, 'tbr', '\t%dk'),
3449 shorten_protocol_name(f
.get('protocol', '')),
3451 format_field(f
, 'vcodec', default
='unknown').replace(
3452 'none', 'images' if f
.get('acodec') == 'none'
3453 else self
._format
_screen
('audio only', self
.Styles
.SUPPRESS
)),
3454 format_field(f
, 'vbr', '\t%dk'),
3455 format_field(f
, 'acodec', default
='unknown').replace(
3456 'none', '' if f
.get('vcodec') == 'none'
3457 else self
._format
_screen
('video only', self
.Styles
.SUPPRESS
)),
3458 format_field(f
, 'abr', '\t%dk'),
3459 format_field(f
, 'asr', '\t%dHz'),
3461 self
._format
_screen
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3462 format_field(f
, 'language', '[%s]'),
3463 join_nonempty(format_field(f
, 'format_note'),
3464 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3467 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3468 header_line
= self
._list
_format
_headers
(
3469 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3470 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3472 return render_table(
3473 header_line
, table
, hide_empty
=True,
3474 delim
=self
._format
_screen
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3476 def render_thumbnails_table(self
, info_dict
):
3477 thumbnails
= list(info_dict
.get('thumbnails'))
3480 return render_table(
3481 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3482 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3484 def render_subtitles_table(self
, video_id
, subtitles
):
3485 def _row(lang
, formats
):
3486 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3487 if len(set(names
)) == 1:
3488 names
= [] if names
[0] == 'unknown' else names
[:1]
3489 return [lang
, ', '.join(names
), ', '.join(exts
)]
3493 return render_table(
3494 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3495 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3498 def __list_table(self
, video_id
, name
, func
, *args
):
3501 self
.to_screen(f
'{video_id} has no {name}')
3503 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3504 self
.to_stdout(table
)
3506 def list_formats(self
, info_dict
):
3507 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3509 def list_thumbnails(self
, info_dict
):
3510 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3512 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3513 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3515 def urlopen(self
, req
):
3516 """ Start an HTTP download """
3517 if isinstance(req
, compat_basestring
):
3518 req
= sanitized_Request(req
)
3519 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3521 def print_debug_header(self
):
3522 if not self
.params
.get('verbose'):
3525 def get_encoding(stream
):
3526 ret
= getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__)
3527 if not supports_terminal_sequences(stream
):
3528 from .compat
import WINDOWS_VT_MODE
3529 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3532 encoding_str
= 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3533 locale
.getpreferredencoding(),
3534 sys
.getfilesystemencoding(),
3535 get_encoding(self
._screen
_file
), get_encoding(self
._err
_file
),
3536 self
.get_encoding())
3538 logger
= self
.params
.get('logger')
3540 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3541 write_debug(encoding_str
)
3543 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3544 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3546 source
= detect_variant()
3547 write_debug(join_nonempty(
3548 'yt-dlp version', __version__
,
3549 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3550 '' if source
== 'unknown' else f
'({source})',
3552 if not _LAZY_LOADER
:
3553 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3554 write_debug('Lazy loading extractors is forcibly disabled')
3556 write_debug('Lazy loading extractors is disabled')
3557 if plugin_extractors
or plugin_postprocessors
:
3558 write_debug('Plugins: %s' % [
3559 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3560 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3561 if self
.params
.get('compat_opts'):
3562 write_debug('Compatibility options: %s' % ', '.join(self
.params
.get('compat_opts')))
3564 if source
== 'source':
3567 ['git', 'rev-parse', '--short', 'HEAD'],
3568 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3569 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3570 out
, err
= sp
.communicate_or_kill()
3571 out
= out
.decode().strip()
3572 if re
.match('[0-9a-f]+', out
):
3573 write_debug('Git HEAD: %s' % out
)
3580 def python_implementation():
3581 impl_name
= platform
.python_implementation()
3582 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3583 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3586 write_debug('Python version %s (%s %s) - %s' % (
3587 platform
.python_version(),
3588 python_implementation(),
3589 platform
.architecture()[0],
3592 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3593 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3595 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features
)
3597 exe_versions
['rtmpdump'] = rtmpdump_version()
3598 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3599 exe_str
= ', '.join(
3600 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3602 write_debug('exe versions: %s' % exe_str
)
3604 from .downloader
.websocket
import has_websockets
3605 from .postprocessor
.embedthumbnail
import has_mutagen
3606 from .cookies
import SQLITE_AVAILABLE
, SECRETSTORAGE_AVAILABLE
3608 lib_str
= join_nonempty(
3609 compat_pycrypto_AES
and compat_pycrypto_AES
.__name
__.split('.')[0],
3610 SECRETSTORAGE_AVAILABLE
and 'secretstorage',
3611 has_mutagen
and 'mutagen',
3612 SQLITE_AVAILABLE
and 'sqlite',
3613 has_websockets
and 'websockets',
3614 delim
=', ') or 'none'
3615 write_debug('Optional libraries: %s' % lib_str
)
3618 for handler
in self
._opener
.handlers
:
3619 if hasattr(handler
, 'proxies'):
3620 proxy_map
.update(handler
.proxies
)
3621 write_debug(f
'Proxy map: {proxy_map}')
3624 if False and self
.params
.get('call_home'):
3625 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3626 write_debug('Public IP address: %s' % ipaddr
)
3627 latest_version
= self
.urlopen(
3628 'https://yt-dl.org/latest/version').read().decode('utf-8')
3629 if version_tuple(latest_version
) > version_tuple(__version__
):
3630 self
.report_warning(
3631 'You are using an outdated version (newest version: %s)! '
3632 'See https://yt-dl.org/update if you need help updating.' %
3635 def _setup_opener(self
):
3636 timeout_val
= self
.params
.get('socket_timeout')
3637 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3639 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3640 opts_cookiefile
= self
.params
.get('cookiefile')
3641 opts_proxy
= self
.params
.get('proxy')
3643 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3645 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3646 if opts_proxy
is not None:
3647 if opts_proxy
== '':
3650 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3652 proxies
= compat_urllib_request
.getproxies()
3653 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3654 if 'http' in proxies
and 'https' not in proxies
:
3655 proxies
['https'] = proxies
['http']
3656 proxy_handler
= PerRequestProxyHandler(proxies
)
3658 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3659 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3660 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3661 redirect_handler
= YoutubeDLRedirectHandler()
3662 data_handler
= compat_urllib_request_DataHandler()
3664 # When passing our own FileHandler instance, build_opener won't add the
3665 # default FileHandler and allows us to disable the file protocol, which
3666 # can be used for malicious purposes (see
3667 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3668 file_handler
= compat_urllib_request
.FileHandler()
3670 def file_open(*args
, **kwargs
):
3671 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3672 file_handler
.file_open
= file_open
3674 opener
= compat_urllib_request
.build_opener(
3675 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3677 # Delete the default user-agent header, which would otherwise apply in
3678 # cases where our custom HTTP handler doesn't come into play
3679 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3680 opener
.addheaders
= []
3681 self
._opener
= opener
3683 def encode(self
, s
):
3684 if isinstance(s
, bytes):
3685 return s
# Already encoded
3688 return s
.encode(self
.get_encoding())
3689 except UnicodeEncodeError as err
:
3690 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3693 def get_encoding(self
):
3694 encoding
= self
.params
.get('encoding')
3695 if encoding
is None:
3696 encoding
= preferredencoding()
3699 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3700 ''' Write infojson and returns True = written, False = skip, None = error '''
3701 if overwrite
is None:
3702 overwrite
= self
.params
.get('overwrites', True)
3703 if not self
.params
.get('writeinfojson'):
3706 self
.write_debug(f
'Skipping writing {label} infojson')
3708 elif not self
._ensure
_dir
_exists
(infofn
):
3710 elif not overwrite
and os
.path
.exists(infofn
):
3711 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3713 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3715 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3716 except (OSError, IOError):
3717 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3721 def _write_description(self
, label
, ie_result
, descfn
):
3722 ''' Write description and returns True = written, False = skip, None = error '''
3723 if not self
.params
.get('writedescription'):
3726 self
.write_debug(f
'Skipping writing {label} description')
3728 elif not self
._ensure
_dir
_exists
(descfn
):
3730 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3731 self
.to_screen(f
'[info] {label.title()} description is already present')
3732 elif ie_result
.get('description') is None:
3733 self
.report_warning(f
'There\'s no {label} description to write')
3737 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3738 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3739 descfile
.write(ie_result
['description'])
3740 except (OSError, IOError):
3741 self
.report_error(f
'Cannot write {label} description file {descfn}')
3745 def _write_subtitles(self
, info_dict
, filename
):
3746 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3748 subtitles
= info_dict
.get('requested_subtitles')
3749 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3750 # subtitles download errors are already managed as troubles in relevant IE
3751 # that way it will silently go on when used with unsupporting IE
3754 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3755 if not sub_filename_base
:
3756 self
.to_screen('[info] Skipping writing video subtitles')
3758 for sub_lang
, sub_info
in subtitles
.items():
3759 sub_format
= sub_info
['ext']
3760 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3761 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3762 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3764 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3765 sub_info
['filepath'] = existing_sub
3766 ret
.append((existing_sub
, sub_filename_final
))
3769 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3770 if sub_info
.get('data') is not None:
3772 # Use newline='' to prevent conversion of newline characters
3773 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3774 with io
.open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3775 subfile
.write(sub_info
['data'])
3776 sub_info
['filepath'] = sub_filename
3777 ret
.append((sub_filename
, sub_filename_final
))
3779 except (OSError, IOError):
3780 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3784 sub_copy
= sub_info
.copy()
3785 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3786 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3787 sub_info
['filepath'] = sub_filename
3788 ret
.append((sub_filename
, sub_filename_final
))
3789 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3790 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3791 raise DownloadError(f
'Unable to download video subtitles for {sub_lang!r}: {err}', err
)
3792 self
.report_warning(f
'Unable to download video subtitles for {sub_lang!r}: {err}')
3795 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3796 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3797 write_all
= self
.params
.get('write_all_thumbnails', False)
3798 thumbnails
, ret
= [], []
3799 if write_all
or self
.params
.get('writethumbnail', False):
3800 thumbnails
= info_dict
.get('thumbnails') or []
3801 multiple
= write_all
and len(thumbnails
) > 1
3803 if thumb_filename_base
is None:
3804 thumb_filename_base
= filename
3805 if thumbnails
and not thumb_filename_base
:
3806 self
.write_debug(f
'Skipping writing {label} thumbnail')
3809 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3810 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3811 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3812 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3813 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3815 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3817 self
.to_screen('[info] %s is already present' % (
3818 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3819 t
['filepath'] = existing_thumb
3820 ret
.append((existing_thumb
, thumb_filename_final
))
3822 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3824 uf
= self
.urlopen(t
['url'])
3825 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3826 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3827 shutil
.copyfileobj(uf
, thumbf
)
3828 ret
.append((thumb_filename
, thumb_filename_final
))
3829 t
['filepath'] = thumb_filename
3830 except network_exceptions
as err
:
3832 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3833 if ret
and not write_all
: