4 from __future__
import absolute_import
, unicode_literals
31 from string
import ascii_letters
35 compat_get_terminal_size
,
42 compat_tokenize_tokenize
,
44 compat_urllib_request
,
45 compat_urllib_request_DataHandler
,
46 windows_enable_vt_mode
,
48 from .cookies
import load_cookies
70 format_decimal_suffix
,
92 PerRequestProxyHandler
,
100 register_socks_protocols
,
101 RejectedVideoReached
,
102 remove_terminal_sequences
,
116 supports_terminal_sequences
,
121 UnavailableVideoError
,
127 YoutubeDLCookieProcessor
,
129 YoutubeDLRedirectHandler
,
131 from .cache
import Cache
132 from .minicurses
import format_text
133 from .extractor
import (
134 gen_extractor_classes
,
137 _PLUGIN_CLASSES
as plugin_extractors
139 from .extractor
.openload
import PhantomJSwrapper
140 from .downloader
import (
142 get_suitable_downloader
,
143 shorten_protocol_name
145 from .downloader
.rtmp
import rtmpdump_version
146 from .postprocessor
import (
149 FFmpegFixupDuplicateMoovPP
,
150 FFmpegFixupDurationPP
,
153 FFmpegFixupStretchedPP
,
154 FFmpegFixupTimestampPP
,
157 MoveFilesAfterDownloadPP
,
158 _PLUGIN_CLASSES
as plugin_postprocessors
160 from .update
import detect_variant
161 from .version
import __version__
, RELEASE_GIT_HEAD
163 if compat_os_name
== 'nt':
167 class YoutubeDL(object):
170 YoutubeDL objects are the ones responsible of downloading the
171 actual video file and writing it to disk if the user has requested
172 it, among some other tasks. In most cases there should be one per
173 program. As, given a video URL, the downloader doesn't know how to
174 extract all the needed information, task that InfoExtractors do, it
175 has to pass the URL to one of them.
177 For this, YoutubeDL objects have a method that allows
178 InfoExtractors to be registered in a given order. When it is passed
179 a URL, the YoutubeDL object handles it to the first InfoExtractor it
180 finds that reports being able to handle it. The InfoExtractor extracts
181 all the information about the video or videos the URL refers to, and
182 YoutubeDL process the extracted information, possibly using a File
183 Downloader to download the video.
185 YoutubeDL objects accept a lot of parameters. In order not to saturate
186 the object constructor with arguments, it receives a dictionary of
187 options instead. These options are available through the params
188 attribute for the InfoExtractors to use. The YoutubeDL also
189 registers itself as the downloader in charge for the InfoExtractors
190 that are added to it, so this is a "mutual registration".
194 username: Username for authentication purposes.
195 password: Password for authentication purposes.
196 videopassword: Password for accessing a video.
197 ap_mso: Adobe Pass multiple-system operator identifier.
198 ap_username: Multiple-system operator account username.
199 ap_password: Multiple-system operator account password.
200 usenetrc: Use netrc for authentication instead.
201 verbose: Print additional info to stdout.
202 quiet: Do not print messages to stdout.
203 no_warnings: Do not print out anything for warnings.
204 forceprint: A dict with keys video/playlist mapped to
205 a list of templates to force print to stdout
206 For compatibility, a single list is also accepted
207 forceurl: Force printing final URL. (Deprecated)
208 forcetitle: Force printing title. (Deprecated)
209 forceid: Force printing ID. (Deprecated)
210 forcethumbnail: Force printing thumbnail URL. (Deprecated)
211 forcedescription: Force printing description. (Deprecated)
212 forcefilename: Force printing final filename. (Deprecated)
213 forceduration: Force printing duration. (Deprecated)
214 forcejson: Force printing info_dict as JSON.
215 dump_single_json: Force printing the info_dict of the whole playlist
216 (or video) as a single JSON line.
217 force_write_download_archive: Force writing download archive regardless
218 of 'skip_download' or 'simulate'.
219 simulate: Do not download the video files. If unset (or None),
220 simulate only if listsubtitles, listformats or list_thumbnails is used
221 format: Video format code. see "FORMAT SELECTION" for more details.
222 You can also pass a function. The function takes 'ctx' as
223 argument and returns the formats to download.
224 See "build_format_selector" for an implementation
225 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
226 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
227 extracting metadata even if the video is not actually
228 available for download (experimental)
229 format_sort: A list of fields by which to sort the video formats.
230 See "Sorting Formats" for more details.
231 format_sort_force: Force the given format_sort. see "Sorting Formats"
233 allow_multiple_video_streams: Allow multiple video streams to be merged
235 allow_multiple_audio_streams: Allow multiple audio streams to be merged
237 check_formats Whether to test if the formats are downloadable.
238 Can be True (check all), False (check none),
239 'selected' (check selected formats),
240 or None (check only if requested by extractor)
241 paths: Dictionary of output paths. The allowed keys are 'home'
242 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
243 outtmpl: Dictionary of templates for output names. Allowed keys
244 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
245 For compatibility with youtube-dl, a single string can also be used
246 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
247 restrictfilenames: Do not allow "&" and spaces in file names
248 trim_file_name: Limit length of filename (extension excluded)
249 windowsfilenames: Force the filenames to be windows compatible
250 ignoreerrors: Do not stop on download/postprocessing errors.
251 Can be 'only_download' to ignore only download errors.
252 Default is 'only_download' for CLI, but False for API
253 skip_playlist_after_errors: Number of allowed failures until the rest of
254 the playlist is skipped
255 force_generic_extractor: Force downloader to use the generic extractor
256 overwrites: Overwrite all video and metadata files if True,
257 overwrite only non-video files if None
258 and don't overwrite any file if False
259 For compatibility with youtube-dl,
260 "nooverwrites" may also be used instead
261 playliststart: Playlist item to start at.
262 playlistend: Playlist item to end at.
263 playlist_items: Specific indices of playlist to download.
264 playlistreverse: Download playlist items in reverse order.
265 playlistrandom: Download playlist items in random order.
266 matchtitle: Download only matching titles.
267 rejecttitle: Reject downloads for matching titles.
268 logger: Log messages to a logging.Logger instance.
269 logtostderr: Log messages to stderr instead of stdout.
270 consoletitle: Display progress in console window's titlebar.
271 writedescription: Write the video description to a .description file
272 writeinfojson: Write the video description to a .info.json file
273 clean_infojson: Remove private fields from the infojson
274 getcomments: Extract video comments. This will not be written to disk
275 unless writeinfojson is also given
276 writeannotations: Write the video annotations to a .annotations.xml file
277 writethumbnail: Write the thumbnail image to a file
278 allow_playlist_files: Whether to write playlists' description, infojson etc
279 also to disk when using the 'write*' options
280 write_all_thumbnails: Write all thumbnail formats to files
281 writelink: Write an internet shortcut file, depending on the
282 current platform (.url/.webloc/.desktop)
283 writeurllink: Write a Windows internet shortcut file (.url)
284 writewebloclink: Write a macOS internet shortcut file (.webloc)
285 writedesktoplink: Write a Linux internet shortcut file (.desktop)
286 writesubtitles: Write the video subtitles to a file
287 writeautomaticsub: Write the automatically generated subtitles to a file
288 allsubtitles: Deprecated - Use subtitleslangs = ['all']
289 Downloads all the subtitles of the video
290 (requires writesubtitles or writeautomaticsub)
291 listsubtitles: Lists all available subtitles for the video
292 subtitlesformat: The format code for subtitles
293 subtitleslangs: List of languages of the subtitles to download (can be regex).
294 The list may contain "all" to refer to all the available
295 subtitles. The language can be prefixed with a "-" to
296 exclude it from the requested languages. Eg: ['all', '-live_chat']
297 keepvideo: Keep the video file after post-processing
298 daterange: A DateRange object, download only if the upload_date is in the range.
299 skip_download: Skip the actual download of the video file
300 cachedir: Location of the cache files in the filesystem.
301 False to disable filesystem cache.
302 noplaylist: Download single video instead of a playlist if in doubt.
303 age_limit: An integer representing the user's age in years.
304 Unsuitable videos for the given age are skipped.
305 min_views: An integer representing the minimum view count the video
306 must have in order to not be skipped.
307 Videos without view count information are always
308 downloaded. None for no limit.
309 max_views: An integer representing the maximum view count.
310 Videos that are more popular than that are not
312 Videos without view count information are always
313 downloaded. None for no limit.
314 download_archive: File name of a file where all downloads are recorded.
315 Videos already present in the file are not downloaded
317 break_on_existing: Stop the download process after attempting to download a
318 file that is in the archive.
319 break_on_reject: Stop the download process when encountering a video that
320 has been filtered out.
321 break_per_url: Whether break_on_reject and break_on_existing
322 should act on each input URL as opposed to for the entire queue
323 cookiefile: File name where cookies should be read from and dumped to
324 cookiesfrombrowser: A tuple containing the name of the browser, the profile
325 name/pathfrom where cookies are loaded, and the name of the
326 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
327 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
328 support RFC 5746 secure renegotiation
329 nocheckcertificate: Do not verify SSL certificates
330 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
331 At the moment, this is only supported by YouTube.
332 proxy: URL of the proxy server to use
333 geo_verification_proxy: URL of the proxy to use for IP address verification
334 on geo-restricted sites.
335 socket_timeout: Time to wait for unresponsive hosts, in seconds
336 bidi_workaround: Work around buggy terminals without bidirectional text
337 support, using fridibi
338 debug_printtraffic:Print out sent and received HTTP traffic
339 include_ads: Download ads as well (deprecated)
340 default_search: Prepend this string if an input url is not valid.
341 'auto' for elaborate guessing
342 encoding: Use this encoding instead of the system-specified.
343 extract_flat: Do not resolve URLs, return the immediate result.
344 Pass in 'in_playlist' to only show this behavior for
346 wait_for_video: If given, wait for scheduled streams to become available.
347 The value should be a tuple containing the range
348 (min_secs, max_secs) to wait between retries
349 postprocessors: A list of dictionaries, each with an entry
350 * key: The name of the postprocessor. See
351 yt_dlp/postprocessor/__init__.py for a list.
352 * when: When to run the postprocessor. Can be one of
353 pre_process|before_dl|post_process|after_move.
354 Assumed to be 'post_process' if not given
355 post_hooks: Deprecated - Register a custom postprocessor instead
356 A list of functions that get called as the final step
357 for each video file, after all postprocessors have been
358 called. The filename will be passed as the only argument.
359 progress_hooks: A list of functions that get called on download
360 progress, with a dictionary with the entries
361 * status: One of "downloading", "error", or "finished".
362 Check this first and ignore unknown values.
363 * info_dict: The extracted info_dict
365 If status is one of "downloading", or "finished", the
366 following properties may also be present:
367 * filename: The final filename (always present)
368 * tmpfilename: The filename we're currently writing to
369 * downloaded_bytes: Bytes on disk
370 * total_bytes: Size of the whole file, None if unknown
371 * total_bytes_estimate: Guess of the eventual file size,
373 * elapsed: The number of seconds since download started.
374 * eta: The estimated time in seconds, None if unknown
375 * speed: The download speed in bytes/second, None if
377 * fragment_index: The counter of the currently
378 downloaded video fragment.
379 * fragment_count: The number of fragments (= individual
380 files that will be merged)
382 Progress hooks are guaranteed to be called at least once
383 (with status "finished") if the download is successful.
384 postprocessor_hooks: A list of functions that get called on postprocessing
385 progress, with a dictionary with the entries
386 * status: One of "started", "processing", or "finished".
387 Check this first and ignore unknown values.
388 * postprocessor: Name of the postprocessor
389 * info_dict: The extracted info_dict
391 Progress hooks are guaranteed to be called at least twice
392 (with status "started" and "finished") if the processing is successful.
393 merge_output_format: Extension to use when merging formats.
394 final_ext: Expected final extension; used to detect when the file was
395 already downloaded and converted
396 fixup: Automatically correct known faults of the file.
398 - "never": do nothing
399 - "warn": only emit a warning
400 - "detect_or_warn": check whether we can do anything
401 about it, warn otherwise (default)
402 source_address: Client-side IP address to bind to.
403 call_home: Boolean, true iff we are allowed to contact the
404 yt-dlp servers for debugging. (BROKEN)
405 sleep_interval_requests: Number of seconds to sleep between requests
407 sleep_interval: Number of seconds to sleep before each download when
408 used alone or a lower bound of a range for randomized
409 sleep before each download (minimum possible number
410 of seconds to sleep) when used along with
412 max_sleep_interval:Upper bound of a range for randomized sleep before each
413 download (maximum possible number of seconds to sleep).
414 Must only be used along with sleep_interval.
415 Actual sleep time will be a random float from range
416 [sleep_interval; max_sleep_interval].
417 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
418 listformats: Print an overview of available video formats and exit.
419 list_thumbnails: Print a table of all thumbnails and exit.
420 match_filter: A function that gets called with the info_dict of
422 If it returns a message, the video is ignored.
423 If it returns None, the video is downloaded.
424 match_filter_func in utils.py is one example for this.
425 no_color: Do not emit color codes in output.
426 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
429 Two-letter ISO 3166-2 country code that will be used for
430 explicit geographic restriction bypassing via faking
431 X-Forwarded-For HTTP header
433 IP range in CIDR notation that will be used similarly to
436 The following options determine which downloader is picked:
437 external_downloader: A dictionary of protocol keys and the executable of the
438 external downloader to use for it. The allowed protocols
439 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
440 Set the value to 'native' to use the native downloader
441 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
442 or {'m3u8': 'ffmpeg'} instead.
443 Use the native HLS downloader instead of ffmpeg/avconv
444 if True, otherwise use ffmpeg/avconv if False, otherwise
445 use downloader suggested by extractor if None.
446 compat_opts: Compatibility options. See "Differences in default behavior".
447 The following options do not work when used through the API:
448 filename, abort-on-error, multistreams, no-live-chat, format-sort
449 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
450 Refer __init__.py for their implementation
451 progress_template: Dictionary of templates for progress outputs.
452 Allowed keys are 'download', 'postprocess',
453 'download-title' (console title) and 'postprocess-title'.
454 The template is mapped on a dictionary with keys 'progress' and 'info'
456 The following parameters are not used by YoutubeDL itself, they are used by
457 the downloader (see yt_dlp/downloader/common.py):
458 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
459 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
460 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
461 external_downloader_args, concurrent_fragment_downloads.
463 The following options are used by the post processors:
464 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
465 otherwise prefer ffmpeg. (avconv support is deprecated)
466 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
467 to the binary or its containing directory.
468 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
469 and a list of additional command-line arguments for the
470 postprocessor/executable. The dict can also have "PP+EXE" keys
471 which are used when the given exe is used by the given PP.
472 Use 'default' as the name for arguments to passed to all PP
473 For compatibility with youtube-dl, a single list of args
476 The following options are used by the extractors:
477 extractor_retries: Number of times to retry for known errors
478 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
479 hls_split_discontinuity: Split HLS playlists to different formats at
480 discontinuities such as ad breaks (default: False)
481 extractor_args: A dictionary of arguments to be passed to the extractors.
482 See "EXTRACTOR ARGUMENTS" for details.
483 Eg: {'youtube': {'skip': ['dash', 'hls']}}
484 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
485 If True (default), DASH manifests and related
486 data will be downloaded and processed by extractor.
487 You can reduce network I/O by disabling it if you don't
488 care about DASH. (only for youtube)
489 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
490 If True (default), HLS manifests and related
491 data will be downloaded and processed by extractor.
492 You can reduce network I/O by disabling it if you don't
493 care about HLS. (only for youtube)
496 _NUMERIC_FIELDS
= set((
497 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
498 'timestamp', 'release_timestamp',
499 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
500 'average_rating', 'comment_count', 'age_limit',
501 'start_time', 'end_time',
502 'chapter_number', 'season_number', 'episode_number',
503 'track_number', 'disc_number', 'release_year',
506 _format_selection_exts
= {
507 'audio': {'m4a', 'mp3', 'ogg', 'aac'}
,
508 'video': {'mp4', 'flv', 'webm', '3gp'}
,
509 'storyboards': {'mhtml'}
,
514 _pps
= {k: [] for k in POSTPROCESS_WHEN}
515 _printed_messages
= set()
516 _first_webpage_request
= True
517 _download_retcode
= None
518 _num_downloads
= None
520 _playlist_urls
= set()
523 def __init__(self
, params
=None, auto_init
=True):
524 """Create a FileDownloader object with the given options.
525 @param auto_init Whether to load the default extractors and print header (if verbose).
526 Set to 'no_verbose_header' to not print the header
531 self
._ies
_instances
= {}
532 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
533 self
._printed
_messages
= set()
534 self
._first
_webpage
_request
= True
535 self
._post
_hooks
= []
536 self
._progress
_hooks
= []
537 self
._postprocessor
_hooks
= []
538 self
._download
_retcode
= 0
539 self
._num
_downloads
= 0
541 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
542 self
._err
_file
= sys
.stderr
544 self
.cache
= Cache(self
)
546 windows_enable_vt_mode()
547 self
._allow
_colors
= {
548 'screen': not self
.params
.get('no_color') and supports_terminal_sequences(self
._screen
_file
),
549 'err': not self
.params
.get('no_color') and supports_terminal_sequences(self
._err
_file
),
552 if sys
.version_info
< (3, 6):
554 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
556 if self
.params
.get('allow_unplayable_formats'):
558 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
559 'This is a developer option intended for debugging. \n'
560 ' If you experience any issues while using this option, '
561 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
563 def check_deprecated(param
, option
, suggestion
):
564 if self
.params
.get(param
) is not None:
565 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
569 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
570 if self
.params
.get('geo_verification_proxy') is None:
571 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
573 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
574 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
575 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
577 for msg
in self
.params
.get('_warnings', []):
578 self
.report_warning(msg
)
579 for msg
in self
.params
.get('_deprecation_warnings', []):
580 self
.deprecation_warning(msg
)
582 if 'list-formats' in self
.params
.get('compat_opts', []):
583 self
.params
['listformats_table'] = False
585 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
586 # nooverwrites was unnecessarily changed to overwrites
587 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
588 # This ensures compatibility with both keys
589 self
.params
['overwrites'] = not self
.params
['nooverwrites']
590 elif self
.params
.get('overwrites') is None:
591 self
.params
.pop('overwrites', None)
593 self
.params
['nooverwrites'] = not self
.params
['overwrites']
595 # Compatibility with older syntax
596 params
.setdefault('forceprint', {})
597 if not isinstance(params
['forceprint'], dict):
598 params
['forceprint'] = {'video': params['forceprint']}
600 if params
.get('bidi_workaround', False):
603 master
, slave
= pty
.openpty()
604 width
= compat_get_terminal_size().columns
608 width_args
= ['-w', str(width
)]
610 stdin
=subprocess
.PIPE
,
612 stderr
=self
._err
_file
)
614 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
616 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
617 self
._output
_channel
= os
.fdopen(master
, 'rb')
618 except OSError as ose
:
619 if ose
.errno
== errno
.ENOENT
:
621 'Could not find fribidi executable, ignoring --bidi-workaround. '
622 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
626 if (sys
.platform
!= 'win32'
627 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
628 and not params
.get('restrictfilenames', False)):
629 # Unicode filesystem API will throw errors (#1474, #13027)
631 'Assuming --restrict-filenames since file system encoding '
632 'cannot encode all characters. '
633 'Set the LC_ALL environment variable to fix this.')
634 self
.params
['restrictfilenames'] = True
636 self
.outtmpl_dict
= self
.parse_outtmpl()
638 # Creating format selector here allows us to catch syntax errors before the extraction
639 self
.format_selector
= (
640 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
641 else self
.params
['format'] if callable(self
.params
['format'])
642 else self
.build_format_selector(self
.params
['format']))
647 if auto_init
!= 'no_verbose_header':
648 self
.print_debug_header()
649 self
.add_default_info_extractors()
652 'post_hooks': self
.add_post_hook
,
653 'progress_hooks': self
.add_progress_hook
,
654 'postprocessor_hooks': self
.add_postprocessor_hook
,
656 for opt
, fn
in hooks
.items():
657 for ph
in self
.params
.get(opt
, []):
660 for pp_def_raw
in self
.params
.get('postprocessors', []):
661 pp_def
= dict(pp_def_raw
)
662 when
= pp_def
.pop('when', 'post_process')
663 self
.add_post_processor(
664 get_postprocessor(pp_def
.pop('key'))(self
, **compat_kwargs(pp_def
)),
667 register_socks_protocols()
669 def preload_download_archive(fn
):
670 """Preload the archive, if any is specified"""
673 self
.write_debug(f
'Loading archive file {fn!r}')
675 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
676 for line
in archive_file
:
677 self
.archive
.add(line
.strip())
678 except IOError as ioe
:
679 if ioe
.errno
!= errno
.ENOENT
:
685 preload_download_archive(self
.params
.get('download_archive'))
687 def warn_if_short_id(self
, argv
):
688 # short YouTube ID starting with dash?
690 i
for i
, a
in enumerate(argv
)
691 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
695 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
696 + ['--'] + [argv
[i
] for i
in idxs
]
699 'Long argument string detected. '
700 'Use -- to separate parameters and URLs, like this:\n%s' %
701 args_to_str(correct_argv
))
703 def add_info_extractor(self
, ie
):
704 """Add an InfoExtractor object to the end of the list."""
706 self
._ies
[ie_key
] = ie
707 if not isinstance(ie
, type):
708 self
._ies
_instances
[ie_key
] = ie
709 ie
.set_downloader(self
)
711 def _get_info_extractor_class(self
, ie_key
):
712 ie
= self
._ies
.get(ie_key
)
714 ie
= get_info_extractor(ie_key
)
715 self
.add_info_extractor(ie
)
718 def get_info_extractor(self
, ie_key
):
720 Get an instance of an IE with name ie_key, it will try to get one from
721 the _ies list, if there's no instance it will create a new one and add
722 it to the extractor list.
724 ie
= self
._ies
_instances
.get(ie_key
)
726 ie
= get_info_extractor(ie_key
)()
727 self
.add_info_extractor(ie
)
730 def add_default_info_extractors(self
):
732 Add the InfoExtractors returned by gen_extractors to the end of the list
734 for ie
in gen_extractor_classes():
735 self
.add_info_extractor(ie
)
737 def add_post_processor(self
, pp
, when
='post_process'):
738 """Add a PostProcessor object to the end of the chain."""
739 self
._pps
[when
].append(pp
)
740 pp
.set_downloader(self
)
742 def add_post_hook(self
, ph
):
743 """Add the post hook"""
744 self
._post
_hooks
.append(ph
)
746 def add_progress_hook(self
, ph
):
747 """Add the download progress hook"""
748 self
._progress
_hooks
.append(ph
)
750 def add_postprocessor_hook(self
, ph
):
751 """Add the postprocessing progress hook"""
752 self
._postprocessor
_hooks
.append(ph
)
753 for pps
in self
._pps
.values():
755 pp
.add_progress_hook(ph
)
757 def _bidi_workaround(self
, message
):
758 if not hasattr(self
, '_output_channel'):
761 assert hasattr(self
, '_output_process')
762 assert isinstance(message
, compat_str
)
763 line_count
= message
.count('\n') + 1
764 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
765 self
._output
_process
.stdin
.flush()
766 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
767 for _
in range(line_count
))
768 return res
[:-len('\n')]
770 def _write_string(self
, message
, out
=None, only_once
=False):
772 if message
in self
._printed
_messages
:
774 self
._printed
_messages
.add(message
)
775 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
777 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
778 """Print message to stdout"""
779 if self
.params
.get('logger'):
780 self
.params
['logger'].debug(message
)
781 elif not quiet
or self
.params
.get('verbose'):
783 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
784 self
._err
_file
if quiet
else self
._screen
_file
)
786 def to_stderr(self
, message
, only_once
=False):
787 """Print message to stderr"""
788 assert isinstance(message
, compat_str
)
789 if self
.params
.get('logger'):
790 self
.params
['logger'].error(message
)
792 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._err
_file
, only_once
=only_once
)
794 def to_console_title(self
, message
):
795 if not self
.params
.get('consoletitle', False):
797 message
= remove_terminal_sequences(message
)
798 if compat_os_name
== 'nt':
799 if ctypes
.windll
.kernel32
.GetConsoleWindow():
800 # c_wchar_p() might not be necessary if `message` is
801 # already of type unicode()
802 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
803 elif 'TERM' in os
.environ
:
804 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
806 def save_console_title(self
):
807 if not self
.params
.get('consoletitle', False):
809 if self
.params
.get('simulate'):
811 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
812 # Save the title on stack
813 self
._write
_string
('\033[22;0t', self
._screen
_file
)
815 def restore_console_title(self
):
816 if not self
.params
.get('consoletitle', False):
818 if self
.params
.get('simulate'):
820 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
821 # Restore the title from stack
822 self
._write
_string
('\033[23;0t', self
._screen
_file
)
825 self
.save_console_title()
828 def __exit__(self
, *args
):
829 self
.restore_console_title()
831 if self
.params
.get('cookiefile') is not None:
832 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
834 def trouble(self
, message
=None, tb
=None, is_error
=True):
835 """Determine action to take when a download problem appears.
837 Depending on if the downloader has been configured to ignore
838 download errors or not, this method may throw an exception or
839 not when errors are found, after printing the message.
841 @param tb If given, is additional traceback information
842 @param is_error Whether to raise error according to ignorerrors
844 if message
is not None:
845 self
.to_stderr(message
)
846 if self
.params
.get('verbose'):
848 if sys
.exc_info()[0]: # if .trouble has been called from an except block
850 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
851 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
852 tb
+= encode_compat_str(traceback
.format_exc())
854 tb_data
= traceback
.format_list(traceback
.extract_stack())
855 tb
= ''.join(tb_data
)
860 if not self
.params
.get('ignoreerrors'):
861 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
862 exc_info
= sys
.exc_info()[1].exc_info
864 exc_info
= sys
.exc_info()
865 raise DownloadError(message
, exc_info
)
866 self
._download
_retcode
= 1
868 def to_screen(self
, message
, skip_eol
=False):
869 """Print message to stdout if not in quiet mode"""
871 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
875 EMPHASIS
= 'light blue'
880 SUPPRESS
= 'light black'
882 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
885 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', 'ascii')
886 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
887 if fallback
is not None and text
!= original_text
:
889 if isinstance(f
, self
.Styles
):
891 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
893 def _format_screen(self
, *args
, **kwargs
):
894 return self
._format
_text
(
895 self
._screen
_file
, self
._allow
_colors
['screen'], *args
, **kwargs
)
897 def _format_err(self
, *args
, **kwargs
):
898 return self
._format
_text
(
899 self
._err
_file
, self
._allow
_colors
['err'], *args
, **kwargs
)
901 def report_warning(self
, message
, only_once
=False):
903 Print the message to stderr, it will be prefixed with 'WARNING:'
904 If stderr is a tty file the 'WARNING:' will be colored
906 if self
.params
.get('logger') is not None:
907 self
.params
['logger'].warning(message
)
909 if self
.params
.get('no_warnings'):
911 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
913 def deprecation_warning(self
, message
):
914 if self
.params
.get('logger') is not None:
915 self
.params
['logger'].warning('DeprecationWarning: {message}')
917 self
.to_stderr(f
'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
919 def report_error(self
, message
, *args
, **kwargs
):
921 Do the same as trouble, but prefixes the message with 'ERROR:', colored
922 in red if stderr is a tty file.
924 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
926 def write_debug(self
, message
, only_once
=False):
927 '''Log debug message or Print message to stderr'''
928 if not self
.params
.get('verbose', False):
930 message
= '[debug] %s' % message
931 if self
.params
.get('logger'):
932 self
.params
['logger'].debug(message
)
934 self
.to_stderr(message
, only_once
)
936 def report_file_already_downloaded(self
, file_name
):
937 """Report file has already been fully downloaded."""
939 self
.to_screen('[download] %s has already been downloaded' % file_name
)
940 except UnicodeEncodeError:
941 self
.to_screen('[download] The file has already been downloaded')
943 def report_file_delete(self
, file_name
):
944 """Report that existing file will be deleted."""
946 self
.to_screen('Deleting existing file %s' % file_name
)
947 except UnicodeEncodeError:
948 self
.to_screen('Deleting existing file')
950 def raise_no_formats(self
, info
, forced
=False):
951 has_drm
= info
.get('__has_drm')
952 msg
= 'This video is DRM protected' if has_drm
else 'No video formats found!'
953 expected
= self
.params
.get('ignore_no_formats_error')
954 if forced
or not expected
:
955 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
956 expected
=has_drm
or expected
)
958 self
.report_warning(msg
)
960 def parse_outtmpl(self
):
961 outtmpl_dict
= self
.params
.get('outtmpl', {})
962 if not isinstance(outtmpl_dict
, dict):
963 outtmpl_dict
= {'default': outtmpl_dict}
964 # Remove spaces in the default template
965 if self
.params
.get('restrictfilenames'):
966 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
968 sanitize
= lambda x
: x
969 outtmpl_dict
.update({
970 k
: sanitize(v
) for k
, v
in DEFAULT_OUTTMPL
.items()
971 if outtmpl_dict
.get(k
) is None})
972 for key
, val
in outtmpl_dict
.items():
973 if isinstance(val
, bytes):
975 'Parameter outtmpl is bytes, but should be a unicode string. '
976 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
979 def get_output_path(self
, dir_type
='', filename
=None):
980 paths
= self
.params
.get('paths', {})
981 assert isinstance(paths
, dict)
983 expand_path(paths
.get('home', '').strip()),
984 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
987 # Temporary fix for #4787
988 # 'Treat' all problem characters by passing filename through preferredencoding
989 # to workaround encoding issues with subprocess on python2 @ Windows
990 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
991 path
= encodeFilename(path
, True).decode(preferredencoding())
992 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
995 def _outtmpl_expandpath(outtmpl
):
996 # expand_path translates '%%' into '%' and '$$' into '$'
997 # correspondingly that is not what we want since we need to keep
998 # '%%' intact for template dict substitution step. Working around
999 # with boundary-alike separator hack.
1000 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
1001 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
1003 # outtmpl should be expand_path'ed before template dict substitution
1004 # because meta fields may contain env variables we don't want to
1005 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1006 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1007 return expand_path(outtmpl
).replace(sep
, '')
1010 def escape_outtmpl(outtmpl
):
1011 ''' Escape any remaining strings like %s, %abc% etc. '''
1013 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1014 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1018 def validate_outtmpl(cls
, outtmpl
):
1019 ''' @return None or Exception object '''
1021 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljqBUDS]'),
1022 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1023 cls
._outtmpl
_expandpath
(outtmpl
))
1025 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1027 except ValueError as err
:
1031 def _copy_infodict(info_dict
):
1032 info_dict
= dict(info_dict
)
1033 for key
in ('__original_infodict', '__postprocessors'):
1034 info_dict
.pop(key
, None)
1037 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1038 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1039 @param sanitize Whether to sanitize the output as a filename.
1040 For backward compatibility, a function can also be passed
1043 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1045 info_dict
= self
._copy
_infodict
(info_dict
)
1046 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1047 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1048 if info_dict
.get('duration', None) is not None
1050 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
1051 info_dict
['video_autonumber'] = self
._num
_videos
1052 if info_dict
.get('resolution') is None:
1053 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1055 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1056 # of %(field)s to %(field)0Nd for backward compatibility
1057 field_size_compat_map
= {
1058 'playlist_index': number_of_digits(info_dict
.get('_last_playlist_index') or 0),
1059 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1060 'autonumber': self
.params
.get('autonumber_size') or 5,
1064 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljqBUDS]'))
1069 # Field is of the form key1.key2...
1070 # where keys (except first) can be string, int or slice
1071 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1072 MATH_FIELD_RE
= r
'''(?:{field}|{num})'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
1073 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1074 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
1077 (?P<maths>(?:{math_op}{math_field})*)
1078 (?:>(?P<strf_format>.+?))?
1079 (?P<alternate>(?<!\\),[^|&)]+)?
1080 (?:&(?P<replacement>.*?))?
1081 (?:\|(?P<default>.*?))?
1082 $'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
1084 def _traverse_infodict(k
):
1088 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1090 def get_value(mdict
):
1092 value
= _traverse_infodict(mdict
['fields'])
1095 value
= float_or_none(value
)
1096 if value
is not None:
1099 offset_key
= mdict
['maths']
1101 value
= float_or_none(value
)
1105 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1106 offset_key
).group(0)
1107 offset_key
= offset_key
[len(item
):]
1108 if operator
is None:
1109 operator
= MATH_FUNCTIONS
[item
]
1111 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1112 offset
= float_or_none(item
)
1114 offset
= float_or_none(_traverse_infodict(item
))
1116 value
= operator(value
, multiplier
* offset
)
1117 except (TypeError, ZeroDivisionError):
1120 # Datetime formatting
1121 if mdict
['strf_format']:
1122 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1126 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1128 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1129 return sanitize_filename(str(value
), restricted
=restricted
,
1130 is_id
=re
.search(r
'(^|[_.])id(\.|$)', key
))
1132 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1133 sanitize
= bool(sanitize
)
1135 def _dumpjson_default(obj
):
1136 if isinstance(obj
, (set, LazyList
)):
1140 def create_key(outer_mobj
):
1141 if not outer_mobj
.group('has_key'):
1142 return outer_mobj
.group(0)
1143 key
= outer_mobj
.group('key')
1144 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1145 initial_field
= mobj
.group('fields') if mobj
else ''
1146 value
, replacement
, default
= None, None, na
1148 mobj
= mobj
.groupdict()
1149 default
= mobj
['default'] if mobj
['default'] is not None else default
1150 value
= get_value(mobj
)
1151 replacement
= mobj
['replacement']
1152 if value
is None and mobj
['alternate']:
1153 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['alternate'][1:])
1157 fmt
= outer_mobj
.group('format')
1158 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1159 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
1161 value
= default
if value
is None else value
if replacement
is None else replacement
1163 flags
= outer_mobj
.group('conversion') or ''
1164 str_fmt
= f
'{fmt[:-1]}s'
1165 if fmt
[-1] == 'l': # list
1166 delim
= '\n' if '#' in flags
else ', '
1167 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1168 elif fmt
[-1] == 'j': # json
1169 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1170 elif fmt
[-1] == 'q': # quoted
1171 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1172 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1173 elif fmt
[-1] == 'B': # bytes
1174 value
= f
'%{str_fmt}'.encode('utf-8') % str(value
).encode('utf-8')
1175 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1176 elif fmt
[-1] == 'U': # unicode normalized
1177 value
, fmt
= unicodedata
.normalize(
1178 # "+" = compatibility equivalence, "#" = NFD
1179 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1181 elif fmt
[-1] == 'D': # decimal suffix
1182 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1183 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1184 factor
=1024 if '#' in flags
else 1000)
1185 elif fmt
[-1] == 'S': # filename sanitization
1186 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1187 elif fmt
[-1] == 'c':
1189 value
= str(value
)[0]
1192 elif fmt
[-1] not in 'rs': # numeric
1193 value
= float_or_none(value
)
1195 value
, fmt
= default
, 's'
1199 # If value is an object, sanitize might convert it to a string
1200 # So we convert it to repr first
1201 value
, fmt
= repr(value
), str_fmt
1202 if fmt
[-1] in 'csr':
1203 value
= sanitizer(initial_field
, value
)
1205 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1206 TMPL_DICT
[key
] = value
1207 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1209 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1211 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1212 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1213 return self
.escape_outtmpl(outtmpl
) % info_dict
1215 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
1217 outtmpl
= self
._outtmpl
_expandpath
(self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default']))
1218 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1220 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
1221 if filename
and force_ext
is not None:
1222 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1224 # https://github.com/blackjack4494/youtube-dlc/issues/85
1225 trim_file_name
= self
.params
.get('trim_file_name', False)
1227 no_ext
, *ext
= filename
.rsplit('.', 2)
1228 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1231 except ValueError as err
:
1232 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1235 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
1236 """Generate the output filename."""
1238 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
1239 if not filename
and dir_type
not in ('', 'temp'):
1243 if not self
.params
.get('paths'):
1245 elif filename
== '-':
1246 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1247 elif os
.path
.isabs(filename
):
1248 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1249 if filename
== '-' or not filename
:
1252 return self
.get_output_path(dir_type
, filename
)
1254 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1255 """ Returns None if the file should be downloaded """
1257 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1260 if 'title' in info_dict
:
1261 # This can happen when we're just evaluating the playlist
1262 title
= info_dict
['title']
1263 matchtitle
= self
.params
.get('matchtitle', False)
1265 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1266 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1267 rejecttitle
= self
.params
.get('rejecttitle', False)
1269 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1270 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1271 date
= info_dict
.get('upload_date')
1272 if date
is not None:
1273 dateRange
= self
.params
.get('daterange', DateRange())
1274 if date
not in dateRange
:
1275 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1276 view_count
= info_dict
.get('view_count')
1277 if view_count
is not None:
1278 min_views
= self
.params
.get('min_views')
1279 if min_views
is not None and view_count
< min_views
:
1280 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1281 max_views
= self
.params
.get('max_views')
1282 if max_views
is not None and view_count
> max_views
:
1283 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1284 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1285 return 'Skipping "%s" because it is age restricted' % video_title
1287 match_filter
= self
.params
.get('match_filter')
1288 if match_filter
is not None:
1290 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1292 # For backward compatibility
1293 ret
= None if incomplete
else match_filter(info_dict
)
1298 if self
.in_download_archive(info_dict
):
1299 reason
= '%s has already been recorded in the archive' % video_title
1300 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1302 reason
= check_filter()
1303 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1304 if reason
is not None:
1306 self
.to_screen('[download] ' + reason
)
1307 if self
.params
.get(break_opt
, False):
1312 def add_extra_info(info_dict
, extra_info
):
1313 '''Set the keys from extra_info in info dict if they are missing'''
1314 for key
, value
in extra_info
.items():
1315 info_dict
.setdefault(key
, value
)
1317 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1318 process
=True, force_generic_extractor
=False):
1320 Return a list with a dictionary for each video extracted.
1323 url -- URL to extract
1326 download -- whether to download videos during extraction
1327 ie_key -- extractor key hint
1328 extra_info -- dictionary containing the extra values to add to each result
1329 process -- whether to resolve all unresolved references (URLs, playlist items),
1330 must be True for download to work.
1331 force_generic_extractor -- force using the generic extractor
1334 if extra_info
is None:
1337 if not ie_key
and force_generic_extractor
:
1341 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1345 for ie_key
, ie
in ies
.items():
1346 if not ie
.suitable(url
):
1349 if not ie
.working():
1350 self
.report_warning('The program functionality for this site has been marked as broken, '
1351 'and will probably not work.')
1353 temp_id
= ie
.get_temp_id(url
)
1354 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1355 self
.to_screen(f
'[{ie_key}] {temp_id}: has already been recorded in the archive')
1356 if self
.params
.get('break_on_existing', False):
1357 raise ExistingVideoReached()
1359 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1361 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1363 def __handle_extraction_exceptions(func
):
1364 @functools.wraps(func
)
1365 def wrapper(self
, *args
, **kwargs
):
1368 return func(self
, *args
, **kwargs
)
1369 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1371 except ReExtractInfo
as e
:
1373 self
.to_screen(f
'{e}; Re-extracting data')
1375 self
.to_stderr('\r')
1376 self
.report_warning(f
'{e}; Re-extracting data')
1378 except GeoRestrictedError
as e
:
1381 msg
+= '\nThis video is available in %s.' % ', '.join(
1382 map(ISO3166Utils
.short2full
, e
.countries
))
1383 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1384 self
.report_error(msg
)
1385 except ExtractorError
as e
: # An error we somewhat expected
1386 self
.report_error(str(e
), e
.format_traceback())
1387 except Exception as e
:
1388 if self
.params
.get('ignoreerrors'):
1389 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1395 def _wait_for_video(self
, ie_result
):
1396 if (not self
.params
.get('wait_for_video')
1397 or ie_result
.get('_type', 'video') != 'video'
1398 or ie_result
.get('formats') or ie_result
.get('url')):
1401 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1406 self
.to_screen(msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r', skip_eol
=True)
1409 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1410 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1411 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1412 diff
= random
.randrange(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
)
1413 self
.report_warning('Release time of video is not known')
1414 elif (diff
or 0) <= 0:
1415 self
.report_warning('Video should already be available according to extracted info')
1416 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1417 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1419 wait_till
= time
.time() + diff
1422 diff
= wait_till
- time
.time()
1425 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1426 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1428 except KeyboardInterrupt:
1430 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1431 except BaseException
as e
:
1432 if not isinstance(e
, ReExtractInfo
):
1436 @__handle_extraction_exceptions
1437 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1438 ie_result
= ie
.extract(url
)
1439 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1441 if isinstance(ie_result
, list):
1442 # Backwards compatibility: old IE result format
1444 '_type': 'compat_list',
1445 'entries': ie_result
,
1447 if extra_info
.get('original_url'):
1448 ie_result
.setdefault('original_url', extra_info
['original_url'])
1449 self
.add_default_extra_info(ie_result
, ie
, url
)
1451 self
._wait
_for
_video
(ie_result
)
1452 return self
.process_ie_result(ie_result
, download
, extra_info
)
1456 def add_default_extra_info(self
, ie_result
, ie
, url
):
1458 self
.add_extra_info(ie_result
, {
1460 'original_url': url
,
1461 'webpage_url_basename': url_basename(url
),
1462 'webpage_url_domain': get_domain(url
),
1465 self
.add_extra_info(ie_result
, {
1466 'extractor': ie
.IE_NAME
,
1467 'extractor_key': ie
.ie_key(),
1470 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1472 Take the result of the ie(may be modified) and resolve all unresolved
1473 references (URLs, playlist items).
1475 It will also download the videos if 'download'.
1476 Returns the resolved ie_result.
1478 if extra_info
is None:
1480 result_type
= ie_result
.get('_type', 'video')
1482 if result_type
in ('url', 'url_transparent'):
1483 ie_result
['url'] = sanitize_url(ie_result
['url'])
1484 if ie_result
.get('original_url'):
1485 extra_info
.setdefault('original_url', ie_result
['original_url'])
1487 extract_flat
= self
.params
.get('extract_flat', False)
1488 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1489 or extract_flat
is True):
1490 info_copy
= ie_result
.copy()
1491 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1492 if ie
and not ie_result
.get('id'):
1493 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1494 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1495 self
.add_extra_info(info_copy
, extra_info
)
1496 info_copy
, _
= self
.pre_process(info_copy
)
1497 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1498 if self
.params
.get('force_write_download_archive', False):
1499 self
.record_download_archive(info_copy
)
1502 if result_type
== 'video':
1503 self
.add_extra_info(ie_result
, extra_info
)
1504 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1505 additional_urls
= (ie_result
or {}).get('additional_urls')
1507 # TODO: Improve MetadataParserPP to allow setting a list
1508 if isinstance(additional_urls
, compat_str
):
1509 additional_urls
= [additional_urls
]
1511 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1512 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1513 ie_result
['additional_entries'] = [
1515 url
, download
, extra_info
=extra_info
,
1516 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1517 for url
in additional_urls
1520 elif result_type
== 'url':
1521 # We have to add extra_info to the results because it may be
1522 # contained in a playlist
1523 return self
.extract_info(
1524 ie_result
['url'], download
,
1525 ie_key
=ie_result
.get('ie_key'),
1526 extra_info
=extra_info
)
1527 elif result_type
== 'url_transparent':
1528 # Use the information from the embedding page
1529 info
= self
.extract_info(
1530 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1531 extra_info
=extra_info
, download
=False, process
=False)
1533 # extract_info may return None when ignoreerrors is enabled and
1534 # extraction failed with an error, don't crash and return early
1539 force_properties
= dict(
1540 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1541 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1542 if f
in force_properties
:
1543 del force_properties
[f
]
1544 new_result
= info
.copy()
1545 new_result
.update(force_properties
)
1547 # Extracted info may not be a video result (i.e.
1548 # info.get('_type', 'video') != video) but rather an url or
1549 # url_transparent. In such cases outer metadata (from ie_result)
1550 # should be propagated to inner one (info). For this to happen
1551 # _type of info should be overridden with url_transparent. This
1552 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1553 if new_result
.get('_type') == 'url':
1554 new_result
['_type'] = 'url_transparent'
1556 return self
.process_ie_result(
1557 new_result
, download
=download
, extra_info
=extra_info
)
1558 elif result_type
in ('playlist', 'multi_video'):
1559 # Protect from infinite recursion due to recursively nested playlists
1560 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1561 webpage_url
= ie_result
['webpage_url']
1562 if webpage_url
in self
._playlist
_urls
:
1564 '[download] Skipping already downloaded playlist: %s'
1565 % ie_result
.get('title') or ie_result
.get('id'))
1568 self
._playlist
_level
+= 1
1569 self
._playlist
_urls
.add(webpage_url
)
1570 self
._sanitize
_thumbnails
(ie_result
)
1572 return self
.__process
_playlist
(ie_result
, download
)
1574 self
._playlist
_level
-= 1
1575 if not self
._playlist
_level
:
1576 self
._playlist
_urls
.clear()
1577 elif result_type
== 'compat_list':
1578 self
.report_warning(
1579 'Extractor %s returned a compat_list result. '
1580 'It needs to be updated.' % ie_result
.get('extractor'))
1583 self
.add_extra_info(r
, {
1584 'extractor': ie_result
['extractor'],
1585 'webpage_url': ie_result
['webpage_url'],
1586 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1587 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1588 'extractor_key': ie_result
['extractor_key'],
1591 ie_result
['entries'] = [
1592 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1593 for r
in ie_result
['entries']
1597 raise Exception('Invalid result type: %s' % result_type
)
1599 def _ensure_dir_exists(self
, path
):
1600 return make_dir(path
, self
.report_error
)
1603 def _playlist_infodict(ie_result
, **kwargs
):
1606 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1607 'playlist_id': ie_result
.get('id'),
1608 'playlist_title': ie_result
.get('title'),
1609 'playlist_uploader': ie_result
.get('uploader'),
1610 'playlist_uploader_id': ie_result
.get('uploader_id'),
1611 'playlist_index': 0,
1615 def __process_playlist(self
, ie_result
, download
):
1616 # We process each entry in the playlist
1617 playlist
= ie_result
.get('title') or ie_result
.get('id')
1618 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1620 if 'entries' not in ie_result
:
1621 raise EntryNotInPlaylist('There are no entries')
1623 MissingEntry
= object()
1624 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1625 if incomplete_entries
:
1626 def fill_missing_entries(entries
, indices
):
1627 ret
= [MissingEntry
] * max(indices
)
1628 for i
, entry
in zip(indices
, entries
):
1631 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1633 playlist_results
= []
1635 playliststart
= self
.params
.get('playliststart', 1)
1636 playlistend
= self
.params
.get('playlistend')
1637 # For backwards compatibility, interpret -1 as whole list
1638 if playlistend
== -1:
1641 playlistitems_str
= self
.params
.get('playlist_items')
1642 playlistitems
= None
1643 if playlistitems_str
is not None:
1644 def iter_playlistitems(format
):
1645 for string_segment
in format
.split(','):
1646 if '-' in string_segment
:
1647 start
, end
= string_segment
.split('-')
1648 for item
in range(int(start
), int(end
) + 1):
1651 yield int(string_segment
)
1652 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1654 ie_entries
= ie_result
['entries']
1655 if isinstance(ie_entries
, list):
1656 playlist_count
= len(ie_entries
)
1657 msg
= f
'Collected {playlist_count} videos; downloading %d of them'
1658 ie_result
['playlist_count'] = ie_result
.get('playlist_count') or playlist_count
1661 return ie_entries
[i
- 1]
1663 msg
= 'Downloading %d videos'
1664 if not isinstance(ie_entries
, (PagedList
, LazyList
)):
1665 ie_entries
= LazyList(ie_entries
)
1666 elif isinstance(ie_entries
, InAdvancePagedList
):
1667 if ie_entries
._pagesize
== 1:
1668 playlist_count
= ie_entries
._pagecount
1671 return YoutubeDL
.__handle
_extraction
_exceptions
(
1672 lambda self
, i
: ie_entries
[i
- 1]
1675 entries
, broken
= [], False
1676 items
= playlistitems
if playlistitems
is not None else itertools
.count(playliststart
)
1680 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1684 entry
= get_entry(i
)
1685 if entry
is MissingEntry
:
1686 raise EntryNotInPlaylist()
1687 except (IndexError, EntryNotInPlaylist
):
1688 if incomplete_entries
:
1689 raise EntryNotInPlaylist(f
'Entry {i} cannot be found')
1690 elif not playlistitems
:
1692 entries
.append(entry
)
1694 if entry
is not None:
1695 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1696 except (ExistingVideoReached
, RejectedVideoReached
):
1699 ie_result
['entries'] = entries
1701 # Save playlist_index before re-ordering
1703 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1704 for i
, entry
in enumerate(entries
, 1)
1705 if entry
is not None]
1706 n_entries
= len(entries
)
1708 if not (ie_result
.get('playlist_count') or broken
or playlistitems
or playlistend
):
1709 ie_result
['playlist_count'] = n_entries
1711 if not playlistitems
and (playliststart
!= 1 or playlistend
):
1712 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1713 ie_result
['requested_entries'] = playlistitems
1715 _infojson_written
= False
1716 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1717 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1718 self
.list_thumbnails(ie_result
)
1719 if write_playlist_files
and not self
.params
.get('simulate'):
1720 ie_copy
= self
._playlist
_infodict
(ie_result
, n_entries
=n_entries
)
1721 _infojson_written
= self
._write
_info
_json
(
1722 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1723 if _infojson_written
is None:
1725 if self
._write
_description
('playlist', ie_result
,
1726 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1728 # TODO: This should be passed to ThumbnailsConvertor if necessary
1729 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1731 if self
.params
.get('playlistreverse', False):
1732 entries
= entries
[::-1]
1733 if self
.params
.get('playlistrandom', False):
1734 random
.shuffle(entries
)
1736 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1738 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1740 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1741 for i
, entry_tuple
in enumerate(entries
, 1):
1742 playlist_index
, entry
= entry_tuple
1743 if 'playlist-index' in self
.params
.get('compat_opts', []):
1744 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1
1745 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1746 # This __x_forwarded_for_ip thing is a bit ugly but requires
1749 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1751 'n_entries': n_entries
,
1752 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1753 'playlist_count': ie_result
.get('playlist_count'),
1754 'playlist_index': playlist_index
,
1755 'playlist_autonumber': i
,
1756 'playlist': playlist
,
1757 'playlist_id': ie_result
.get('id'),
1758 'playlist_title': ie_result
.get('title'),
1759 'playlist_uploader': ie_result
.get('uploader'),
1760 'playlist_uploader_id': ie_result
.get('uploader_id'),
1761 'extractor': ie_result
['extractor'],
1762 'webpage_url': ie_result
['webpage_url'],
1763 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1764 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1765 'extractor_key': ie_result
['extractor_key'],
1768 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1771 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1772 if not entry_result
:
1774 if failures
>= max_failures
:
1776 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1778 playlist_results
.append(entry_result
)
1779 ie_result
['entries'] = playlist_results
1781 # Write the updated info to json
1782 if _infojson_written
and self
._write
_info
_json
(
1783 'updated playlist', ie_result
,
1784 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1787 ie_result
= self
.run_all_pps('playlist', ie_result
)
1788 self
.to_screen(f
'[download] Finished downloading playlist: {playlist}')
1791 @__handle_extraction_exceptions
1792 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1793 return self
.process_ie_result(
1794 entry
, download
=download
, extra_info
=extra_info
)
1796 def _build_format_filter(self
, filter_spec
):
1797 " Returns a function to filter the formats according to the filter_spec "
1807 operator_rex
= re
.compile(r
'''(?x)\s*
1808 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1809 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1810 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1811 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1812 m
= operator_rex
.fullmatch(filter_spec
)
1815 comparison_value
= int(m
.group('value'))
1817 comparison_value
= parse_filesize(m
.group('value'))
1818 if comparison_value
is None:
1819 comparison_value
= parse_filesize(m
.group('value') + 'B')
1820 if comparison_value
is None:
1822 'Invalid value %r in format specification %r' % (
1823 m
.group('value'), filter_spec
))
1824 op
= OPERATORS
[m
.group('op')]
1829 '^=': lambda attr
, value
: attr
.startswith(value
),
1830 '$=': lambda attr
, value
: attr
.endswith(value
),
1831 '*=': lambda attr
, value
: value
in attr
,
1833 str_operator_rex
= re
.compile(r
'''(?x)\s*
1834 (?P<key>[a-zA-Z0-9._-]+)\s*
1835 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1836 (?P<value>[a-zA-Z0-9._-]+)\s*
1837 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1838 m
= str_operator_rex
.fullmatch(filter_spec
)
1840 comparison_value
= m
.group('value')
1841 str_op
= STR_OPERATORS
[m
.group('op')]
1842 if m
.group('negation'):
1843 op
= lambda attr
, value
: not str_op(attr
, value
)
1848 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1851 actual_value
= f
.get(m
.group('key'))
1852 if actual_value
is None:
1853 return m
.group('none_inclusive')
1854 return op(actual_value
, comparison_value
)
1857 def _check_formats(self
, formats
):
1859 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1860 path
= self
.get_output_path('temp')
1861 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1863 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1866 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1867 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1870 if os
.path
.exists(temp_file
.name
):
1872 os
.remove(temp_file
.name
)
1874 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1878 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1880 def _default_format_spec(self
, info_dict
, download
=True):
1883 merger
= FFmpegMergerPP(self
)
1884 return merger
.available
and merger
.can_merge()
1887 not self
.params
.get('simulate')
1891 or info_dict
.get('is_live', False)
1892 or self
.outtmpl_dict
['default'] == '-'))
1895 or self
.params
.get('allow_multiple_audio_streams', False)
1896 or 'format-spec' in self
.params
.get('compat_opts', []))
1899 'best/bestvideo+bestaudio' if prefer_best
1900 else 'bestvideo*+bestaudio/best' if not compat
1901 else 'bestvideo+bestaudio/best')
1903 def build_format_selector(self
, format_spec
):
1904 def syntax_error(note
, start
):
1906 'Invalid format specification: '
1907 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1908 return SyntaxError(message
)
1910 PICKFIRST
= 'PICKFIRST'
1914 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1916 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1917 'video': self
.params
.get('allow_multiple_video_streams', False)}
1919 check_formats
= self
.params
.get('check_formats') == 'selected'
1921 def _parse_filter(tokens
):
1923 for type, string
, start
, _
, _
in tokens
:
1924 if type == tokenize
.OP
and string
== ']':
1925 return ''.join(filter_parts
)
1927 filter_parts
.append(string
)
1929 def _remove_unused_ops(tokens
):
1930 # Remove operators that we don't use and join them with the surrounding strings
1931 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1932 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1933 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1934 for type, string
, start
, end
, line
in tokens
:
1935 if type == tokenize
.OP
and string
== '[':
1937 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1939 yield type, string
, start
, end
, line
1940 # everything inside brackets will be handled by _parse_filter
1941 for type, string
, start
, end
, line
in tokens
:
1942 yield type, string
, start
, end
, line
1943 if type == tokenize
.OP
and string
== ']':
1945 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1947 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1949 yield type, string
, start
, end
, line
1950 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1952 last_string
= string
1956 last_string
+= string
1958 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1960 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1962 current_selector
= None
1963 for type, string
, start
, _
, _
in tokens
:
1964 # ENCODING is only defined in python 3.x
1965 if type == getattr(tokenize
, 'ENCODING', None):
1967 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1968 current_selector
= FormatSelector(SINGLE
, string
, [])
1969 elif type == tokenize
.OP
:
1971 if not inside_group
:
1972 # ')' will be handled by the parentheses group
1973 tokens
.restore_last_token()
1975 elif inside_merge
and string
in ['/', ',']:
1976 tokens
.restore_last_token()
1978 elif inside_choice
and string
== ',':
1979 tokens
.restore_last_token()
1982 if not current_selector
:
1983 raise syntax_error('"," must follow a format selector', start
)
1984 selectors
.append(current_selector
)
1985 current_selector
= None
1987 if not current_selector
:
1988 raise syntax_error('"/" must follow a format selector', start
)
1989 first_choice
= current_selector
1990 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1991 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1993 if not current_selector
:
1994 current_selector
= FormatSelector(SINGLE
, 'best', [])
1995 format_filter
= _parse_filter(tokens
)
1996 current_selector
.filters
.append(format_filter
)
1998 if current_selector
:
1999 raise syntax_error('Unexpected "("', start
)
2000 group
= _parse_format_selection(tokens
, inside_group
=True)
2001 current_selector
= FormatSelector(GROUP
, group
, [])
2003 if not current_selector
:
2004 raise syntax_error('Unexpected "+"', start
)
2005 selector_1
= current_selector
2006 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2008 raise syntax_error('Expected a selector', start
)
2009 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2011 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
2012 elif type == tokenize
.ENDMARKER
:
2014 if current_selector
:
2015 selectors
.append(current_selector
)
2018 def _merge(formats_pair
):
2019 format_1
, format_2
= formats_pair
2022 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2023 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2025 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2026 get_no_more
= {'video': False, 'audio': False}
2027 for (i
, fmt_info
) in enumerate(formats_info
):
2028 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2031 for aud_vid
in ['audio', 'video']:
2032 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2033 if get_no_more
[aud_vid
]:
2036 get_no_more
[aud_vid
] = True
2038 if len(formats_info
) == 1:
2039 return formats_info
[0]
2041 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2042 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2044 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2045 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2047 output_ext
= self
.params
.get('merge_output_format')
2050 output_ext
= the_only_video
['ext']
2051 elif the_only_audio
and not video_fmts
:
2052 output_ext
= the_only_audio
['ext']
2056 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2059 'requested_formats': formats_info
,
2060 'format': '+'.join(filtered('format')),
2061 'format_id': '+'.join(filtered('format_id')),
2063 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2064 'language': '+'.join(orderedSet(filtered('language'))) or None,
2065 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2066 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2067 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2072 'width': the_only_video
.get('width'),
2073 'height': the_only_video
.get('height'),
2074 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2075 'fps': the_only_video
.get('fps'),
2076 'dynamic_range': the_only_video
.get('dynamic_range'),
2077 'vcodec': the_only_video
.get('vcodec'),
2078 'vbr': the_only_video
.get('vbr'),
2079 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2084 'acodec': the_only_audio
.get('acodec'),
2085 'abr': the_only_audio
.get('abr'),
2086 'asr': the_only_audio
.get('asr'),
2091 def _check_formats(formats
):
2092 if not check_formats
:
2095 yield from self
._check
_formats
(formats
)
2097 def _build_selector_function(selector
):
2098 if isinstance(selector
, list): # ,
2099 fs
= [_build_selector_function(s
) for s
in selector
]
2101 def selector_function(ctx
):
2104 return selector_function
2106 elif selector
.type == GROUP
: # ()
2107 selector_function
= _build_selector_function(selector
.selector
)
2109 elif selector
.type == PICKFIRST
: # /
2110 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2112 def selector_function(ctx
):
2114 picked_formats
= list(f(ctx
))
2116 return picked_formats
2119 elif selector
.type == MERGE
: # +
2120 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2122 def selector_function(ctx
):
2123 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2126 elif selector
.type == SINGLE
: # atom
2127 format_spec
= selector
.selector
or 'best'
2129 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2130 if format_spec
== 'all':
2131 def selector_function(ctx
):
2132 yield from _check_formats(ctx
['formats'][::-1])
2133 elif format_spec
== 'mergeall':
2134 def selector_function(ctx
):
2135 formats
= list(_check_formats(ctx
['formats']))
2138 merged_format
= formats
[-1]
2139 for f
in formats
[-2::-1]:
2140 merged_format
= _merge((merged_format
, f
))
2144 format_fallback
, format_reverse
, format_idx
= False, True, 1
2146 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2148 if mobj
is not None:
2149 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2150 format_reverse
= mobj
.group('bw')[0] == 'b'
2151 format_type
= (mobj
.group('type') or [None])[0]
2152 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2153 format_modified
= mobj
.group('mod') is not None
2155 format_fallback
= not format_type
and not format_modified
# for b, w
2157 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2158 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2159 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2160 if format_type
# bv, ba, wv, wa
2161 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2162 if not format_modified
# b, w
2163 else lambda f
: True) # b*, w*
2164 filter_f
= lambda f
: _filter_f(f
) and (
2165 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2167 if format_spec
in self
._format
_selection
_exts
['audio']:
2168 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2169 elif format_spec
in self
._format
_selection
_exts
['video']:
2170 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2171 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2172 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2174 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2176 def selector_function(ctx
):
2177 formats
= list(ctx
['formats'])
2178 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2179 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
2180 # for extractors with incomplete formats (audio only (soundcloud)
2181 # or video only (imgur)) best/worst will fallback to
2182 # best/worst {video,audio}-only format
2184 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2186 yield matches
[format_idx
- 1]
2190 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2192 def final_selector(ctx
):
2193 ctx_copy
= dict(ctx
)
2194 for _filter
in filters
:
2195 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2196 return selector_function(ctx_copy
)
2197 return final_selector
2199 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
2201 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
2202 except tokenize
.TokenError
:
2203 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2205 class TokenIterator(object):
2206 def __init__(self
, tokens
):
2207 self
.tokens
= tokens
2214 if self
.counter
>= len(self
.tokens
):
2215 raise StopIteration()
2216 value
= self
.tokens
[self
.counter
]
2222 def restore_last_token(self
):
2225 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2226 return _build_selector_function(parsed_selector
)
2228 def _calc_headers(self
, info_dict
):
2229 res
= std_headers
.copy()
2231 add_headers
= info_dict
.get('http_headers')
2233 res
.update(add_headers
)
2235 cookies
= self
._calc
_cookies
(info_dict
)
2237 res
['Cookie'] = cookies
2239 if 'X-Forwarded-For' not in res
:
2240 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2241 if x_forwarded_for_ip
:
2242 res
['X-Forwarded-For'] = x_forwarded_for_ip
2246 def _calc_cookies(self
, info_dict
):
2247 pr
= sanitized_Request(info_dict
['url'])
2248 self
.cookiejar
.add_cookie_header(pr
)
2249 return pr
.get_header('Cookie')
2251 def _sort_thumbnails(self
, thumbnails
):
2252 thumbnails
.sort(key
=lambda t
: (
2253 t
.get('preference') if t
.get('preference') is not None else -1,
2254 t
.get('width') if t
.get('width') is not None else -1,
2255 t
.get('height') if t
.get('height') is not None else -1,
2256 t
.get('id') if t
.get('id') is not None else '',
2259 def _sanitize_thumbnails(self
, info_dict
):
2260 thumbnails
= info_dict
.get('thumbnails')
2261 if thumbnails
is None:
2262 thumbnail
= info_dict
.get('thumbnail')
2264 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2268 def check_thumbnails(thumbnails
):
2269 for t
in thumbnails
:
2270 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2272 self
.urlopen(HEADRequest(t
['url']))
2273 except network_exceptions
as err
:
2274 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2278 self
._sort
_thumbnails
(thumbnails
)
2279 for i
, t
in enumerate(thumbnails
):
2280 if t
.get('id') is None:
2282 if t
.get('width') and t
.get('height'):
2283 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2284 t
['url'] = sanitize_url(t
['url'])
2286 if self
.params
.get('check_formats') is True:
2287 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2289 info_dict
['thumbnails'] = thumbnails
2291 def process_video_result(self
, info_dict
, download
=True):
2292 assert info_dict
.get('_type', 'video') == 'video'
2293 self
._num
_videos
+= 1
2295 if 'id' not in info_dict
:
2296 raise ExtractorError('Missing "id" field in extractor result')
2297 if 'title' not in info_dict
:
2298 raise ExtractorError('Missing "title" field in extractor result',
2299 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2301 def report_force_conversion(field
, field_not
, conversion
):
2302 self
.report_warning(
2303 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2304 % (field
, field_not
, conversion
))
2306 def sanitize_string_field(info
, string_field
):
2307 field
= info
.get(string_field
)
2308 if field
is None or isinstance(field
, compat_str
):
2310 report_force_conversion(string_field
, 'a string', 'string')
2311 info
[string_field
] = compat_str(field
)
2313 def sanitize_numeric_fields(info
):
2314 for numeric_field
in self
._NUMERIC
_FIELDS
:
2315 field
= info
.get(numeric_field
)
2316 if field
is None or isinstance(field
, compat_numeric_types
):
2318 report_force_conversion(numeric_field
, 'numeric', 'int')
2319 info
[numeric_field
] = int_or_none(field
)
2321 sanitize_string_field(info_dict
, 'id')
2322 sanitize_numeric_fields(info_dict
)
2324 if 'playlist' not in info_dict
:
2325 # It isn't part of a playlist
2326 info_dict
['playlist'] = None
2327 info_dict
['playlist_index'] = None
2329 self
._sanitize
_thumbnails
(info_dict
)
2331 thumbnail
= info_dict
.get('thumbnail')
2332 thumbnails
= info_dict
.get('thumbnails')
2334 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2336 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2338 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2339 info_dict
['display_id'] = info_dict
['id']
2341 if info_dict
.get('duration') is not None:
2342 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2344 for ts_key
, date_key
in (
2345 ('timestamp', 'upload_date'),
2346 ('release_timestamp', 'release_date'),
2347 ('modified_timestamp', 'modified_date'),
2349 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2350 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2351 # see http://bugs.python.org/issue1646728)
2353 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2354 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2355 except (ValueError, OverflowError, OSError):
2358 live_keys
= ('is_live', 'was_live')
2359 live_status
= info_dict
.get('live_status')
2360 if live_status
is None:
2361 for key
in live_keys
:
2362 if info_dict
.get(key
) is False:
2364 if info_dict
.get(key
):
2367 if all(info_dict
.get(key
) is False for key
in live_keys
):
2368 live_status
= 'not_live'
2370 info_dict
['live_status'] = live_status
2371 for key
in live_keys
:
2372 if info_dict
.get(key
) is None:
2373 info_dict
[key
] = (live_status
== key
)
2375 # Auto generate title fields corresponding to the *_number fields when missing
2376 # in order to always have clean titles. This is very common for TV series.
2377 for field
in ('chapter', 'season', 'episode'):
2378 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2379 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2381 for cc_kind
in ('subtitles', 'automatic_captions'):
2382 cc
= info_dict
.get(cc_kind
)
2384 for _
, subtitle
in cc
.items():
2385 for subtitle_format
in subtitle
:
2386 if subtitle_format
.get('url'):
2387 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2388 if subtitle_format
.get('ext') is None:
2389 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2391 automatic_captions
= info_dict
.get('automatic_captions')
2392 subtitles
= info_dict
.get('subtitles')
2394 info_dict
['requested_subtitles'] = self
.process_subtitles(
2395 info_dict
['id'], subtitles
, automatic_captions
)
2397 if info_dict
.get('formats') is None:
2398 # There's only one format available
2399 formats
= [info_dict
]
2401 formats
= info_dict
['formats']
2403 info_dict
['__has_drm'] = any(f
.get('has_drm') for f
in formats
)
2404 if not self
.params
.get('allow_unplayable_formats'):
2405 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2407 # backward compatibility
2408 info_dict
['fulltitle'] = info_dict
['title']
2410 if info_dict
.get('is_live'):
2411 get_from_start
= bool(self
.params
.get('live_from_start'))
2412 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2413 if not get_from_start
:
2414 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2417 self
.raise_no_formats(info_dict
)
2419 def is_wellformed(f
):
2422 self
.report_warning(
2423 '"url" field is missing or empty - skipping format, '
2424 'there is an error in extractor')
2426 if isinstance(url
, bytes):
2427 sanitize_string_field(f
, 'url')
2430 # Filter out malformed formats for better extraction robustness
2431 formats
= list(filter(is_wellformed
, formats
))
2435 # We check that all the formats have the format and format_id fields
2436 for i
, format
in enumerate(formats
):
2437 sanitize_string_field(format
, 'format_id')
2438 sanitize_numeric_fields(format
)
2439 format
['url'] = sanitize_url(format
['url'])
2440 if not format
.get('format_id'):
2441 format
['format_id'] = compat_str(i
)
2443 # Sanitize format_id from characters used in format selector expression
2444 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2445 format_id
= format
['format_id']
2446 if format_id
not in formats_dict
:
2447 formats_dict
[format_id
] = []
2448 formats_dict
[format_id
].append(format
)
2450 # Make sure all formats have unique format_id
2451 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2452 for format_id
, ambiguous_formats
in formats_dict
.items():
2453 ambigious_id
= len(ambiguous_formats
) > 1
2454 for i
, format
in enumerate(ambiguous_formats
):
2456 format
['format_id'] = '%s-%d' % (format_id
, i
)
2457 if format
.get('ext') is None:
2458 format
['ext'] = determine_ext(format
['url']).lower()
2459 # Ensure there is no conflict between id and ext in format selection
2460 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2461 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2462 format
['format_id'] = 'f%s' % format
['format_id']
2464 for i
, format
in enumerate(formats
):
2465 if format
.get('format') is None:
2466 format
['format'] = '{id} - {res}{note}'.format(
2467 id=format
['format_id'],
2468 res
=self
.format_resolution(format
),
2469 note
=format_field(format
, 'format_note', ' (%s)'),
2471 if format
.get('protocol') is None:
2472 format
['protocol'] = determine_protocol(format
)
2473 if format
.get('resolution') is None:
2474 format
['resolution'] = self
.format_resolution(format
, default
=None)
2475 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2476 format
['dynamic_range'] = 'SDR'
2477 if (info_dict
.get('duration') and format
.get('tbr')
2478 and not format
.get('filesize') and not format
.get('filesize_approx')):
2479 format
['filesize_approx'] = info_dict
['duration'] * format
['tbr'] * (1024 / 8)
2481 # Add HTTP headers, so that external programs can use them from the
2483 full_format_info
= info_dict
.copy()
2484 full_format_info
.update(format
)
2485 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2486 # Remove private housekeeping stuff
2487 if '__x_forwarded_for_ip' in info_dict
:
2488 del info_dict
['__x_forwarded_for_ip']
2490 # TODO Central sorting goes here
2492 if self
.params
.get('check_formats') is True:
2493 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2495 if not formats
or formats
[0] is not info_dict
:
2496 # only set the 'formats' fields if the original info_dict list them
2497 # otherwise we end up with a circular reference, the first (and unique)
2498 # element in the 'formats' field in info_dict is info_dict itself,
2499 # which can't be exported to json
2500 info_dict
['formats'] = formats
2502 info_dict
, _
= self
.pre_process(info_dict
)
2504 # The pre-processors may have modified the formats
2505 formats
= info_dict
.get('formats', [info_dict
])
2507 list_only
= self
.params
.get('simulate') is None and (
2508 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2509 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2510 if self
.params
.get('list_thumbnails'):
2511 self
.list_thumbnails(info_dict
)
2512 if self
.params
.get('listsubtitles'):
2513 if 'automatic_captions' in info_dict
:
2514 self
.list_subtitles(
2515 info_dict
['id'], automatic_captions
, 'automatic captions')
2516 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2517 if self
.params
.get('listformats') or interactive_format_selection
:
2518 self
.list_formats(info_dict
)
2520 # Without this printing, -F --print-json will not work
2521 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2524 format_selector
= self
.format_selector
2525 if format_selector
is None:
2526 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2527 self
.write_debug('Default format spec: %s' % req_format
)
2528 format_selector
= self
.build_format_selector(req_format
)
2531 if interactive_format_selection
:
2533 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2535 format_selector
= self
.build_format_selector(req_format
)
2536 except SyntaxError as err
:
2537 self
.report_error(err
, tb
=False, is_error
=False)
2540 # While in format selection we may need to have an access to the original
2541 # format set in order to calculate some metrics or do some processing.
2542 # For now we need to be able to guess whether original formats provided
2543 # by extractor are incomplete or not (i.e. whether extractor provides only
2544 # video-only or audio-only formats) for proper formats selection for
2545 # extractors with such incomplete formats (see
2546 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2547 # Since formats may be filtered during format selection and may not match
2548 # the original formats the results may be incorrect. Thus original formats
2549 # or pre-calculated metrics should be passed to format selection routines
2551 # We will pass a context object containing all necessary additional data
2552 # instead of just formats.
2553 # This fixes incorrect format selection issue (see
2554 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2555 incomplete_formats
= (
2556 # All formats are video-only or
2557 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2558 # all formats are audio-only
2559 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2563 'incomplete_formats': incomplete_formats
,
2566 formats_to_download
= list(format_selector(ctx
))
2567 if interactive_format_selection
and not formats_to_download
:
2568 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2572 if not formats_to_download
:
2573 if not self
.params
.get('ignore_no_formats_error'):
2574 raise ExtractorError('Requested format is not available', expected
=True,
2575 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2576 self
.report_warning('Requested format is not available')
2577 # Process what we can, even without any available formats.
2578 formats_to_download
= [{}]
2580 best_format
= formats_to_download
[-1]
2584 f
'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2585 + ', '.join([f
['format_id'] for f
in formats_to_download
]))
2586 max_downloads_reached
= False
2587 for i
, fmt
in enumerate(formats_to_download
):
2588 formats_to_download
[i
] = new_info
= dict(info_dict
)
2589 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2590 new_info
.update(fmt
)
2591 new_info
['__original_infodict'] = info_dict
2593 self
.process_info(new_info
)
2594 except MaxDownloadsReached
:
2595 max_downloads_reached
= True
2596 new_info
.pop('__original_infodict')
2597 # Remove copied info
2598 for key
, val
in tuple(new_info
.items()):
2599 if info_dict
.get(key
) == val
:
2601 if max_downloads_reached
:
2604 write_archive
= set(f
.get('__write_download_archive', False) for f
in formats_to_download
)
2605 assert write_archive
.issubset({True, False, 'ignore'}
)
2606 if True in write_archive
and False not in write_archive
:
2607 self
.record_download_archive(info_dict
)
2609 info_dict
['requested_downloads'] = formats_to_download
2610 info_dict
= self
.run_all_pps('after_video', info_dict
)
2611 if max_downloads_reached
:
2612 raise MaxDownloadsReached()
2614 # We update the info dict with the selected best quality format (backwards compatibility)
2615 info_dict
.update(best_format
)
2618 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2619 """Select the requested subtitles and their format"""
2621 if normal_subtitles
and self
.params
.get('writesubtitles'):
2622 available_subs
.update(normal_subtitles
)
2623 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2624 for lang
, cap_info
in automatic_captions
.items():
2625 if lang
not in available_subs
:
2626 available_subs
[lang
] = cap_info
2628 if (not self
.params
.get('writesubtitles') and not
2629 self
.params
.get('writeautomaticsub') or not
2633 all_sub_langs
= available_subs
.keys()
2634 if self
.params
.get('allsubtitles', False):
2635 requested_langs
= all_sub_langs
2636 elif self
.params
.get('subtitleslangs', False):
2637 # A list is used so that the order of languages will be the same as
2638 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2639 requested_langs
= []
2640 for lang_re
in self
.params
.get('subtitleslangs'):
2641 if lang_re
== 'all':
2642 requested_langs
.extend(all_sub_langs
)
2644 discard
= lang_re
[0] == '-'
2646 lang_re
= lang_re
[1:]
2647 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2649 for lang
in current_langs
:
2650 while lang
in requested_langs
:
2651 requested_langs
.remove(lang
)
2653 requested_langs
.extend(current_langs
)
2654 requested_langs
= orderedSet(requested_langs
)
2655 elif 'en' in available_subs
:
2656 requested_langs
= ['en']
2658 requested_langs
= [list(all_sub_langs
)[0]]
2660 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2662 formats_query
= self
.params
.get('subtitlesformat', 'best')
2663 formats_preference
= formats_query
.split('/') if formats_query
else []
2665 for lang
in requested_langs
:
2666 formats
= available_subs
.get(lang
)
2668 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2670 for ext
in formats_preference
:
2674 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2680 self
.report_warning(
2681 'No subtitle format found matching "%s" for language %s, '
2682 'using %s' % (formats_query
, lang
, f
['ext']))
2686 def _forceprint(self
, tmpl
, info_dict
):
2687 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2688 if mobj
and mobj
.group(1):
2689 tmpl
= f
'{tmpl[:-1]} = %({tmpl[:-1]})r'
2691 tmpl
= '%({})s'.format(tmpl
)
2693 info_dict
= info_dict
.copy()
2694 info_dict
['formats_table'] = self
.render_formats_table(info_dict
)
2695 info_dict
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2696 info_dict
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2697 info_dict
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2698 self
.to_stdout(self
.evaluate_outtmpl(tmpl
, info_dict
))
2700 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2701 def print_mandatory(field
, actual_field
=None):
2702 if actual_field
is None:
2703 actual_field
= field
2704 if (self
.params
.get('force%s' % field
, False)
2705 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2706 self
.to_stdout(info_dict
[actual_field
])
2708 def print_optional(field
):
2709 if (self
.params
.get('force%s' % field
, False)
2710 and info_dict
.get(field
) is not None):
2711 self
.to_stdout(info_dict
[field
])
2713 info_dict
= info_dict
.copy()
2714 if filename
is not None:
2715 info_dict
['filename'] = filename
2716 if info_dict
.get('requested_formats') is not None:
2717 # For RTMP URLs, also include the playpath
2718 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2719 elif 'url' in info_dict
:
2720 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2722 if self
.params
['forceprint'].get('video') or self
.params
.get('forcejson'):
2723 self
.post_extract(info_dict
)
2724 for tmpl
in self
.params
['forceprint'].get('video', []):
2725 self
._forceprint
(tmpl
, info_dict
)
2727 print_mandatory('title')
2728 print_mandatory('id')
2729 print_mandatory('url', 'urls')
2730 print_optional('thumbnail')
2731 print_optional('description')
2732 print_optional('filename')
2733 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2734 self
.to_stdout(formatSeconds(info_dict
['duration']))
2735 print_mandatory('format')
2737 if self
.params
.get('forcejson'):
2738 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2740 def dl(self
, name
, info
, subtitle
=False, test
=False):
2741 if not info
.get('url'):
2742 self
.raise_no_formats(info
, True)
2745 verbose
= self
.params
.get('verbose')
2748 'quiet': self
.params
.get('quiet') or not verbose
,
2750 'noprogress': not verbose
,
2752 'skip_unavailable_fragments': False,
2753 'keep_fragments': False,
2755 '_no_ytdl_file': True,
2758 params
= self
.params
2759 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2761 for ph
in self
._progress
_hooks
:
2762 fd
.add_progress_hook(ph
)
2764 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2765 for f
in info
.get('requested_formats', []) or [info
])
2766 self
.write_debug('Invoking downloader on "%s"' % urls
)
2768 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2769 # But it may contain objects that are not deep-copyable
2770 new_info
= self
._copy
_infodict
(info
)
2771 if new_info
.get('http_headers') is None:
2772 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2773 return fd
.download(name
, new_info
, subtitle
)
2775 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2776 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2777 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2778 return existing_files
[0]
2780 for file in existing_files
:
2781 self
.report_file_delete(file)
2785 def process_info(self
, info_dict
):
2786 """Process a single resolved IE result. (Modified it in-place)"""
2788 assert info_dict
.get('_type', 'video') == 'video'
2789 original_infodict
= info_dict
2791 if 'format' not in info_dict
and 'ext' in info_dict
:
2792 info_dict
['format'] = info_dict
['ext']
2794 if self
._match
_entry
(info_dict
) is not None:
2795 info_dict
['__write_download_archive'] = 'ignore'
2798 self
.post_extract(info_dict
)
2799 self
._num
_downloads
+= 1
2801 # info_dict['_filename'] needs to be set for backward compatibility
2802 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2803 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2807 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2809 if self
.params
.get('simulate'):
2810 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2813 if full_filename
is None:
2815 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2817 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2820 if self
._write
_description
('video', info_dict
,
2821 self
.prepare_filename(info_dict
, 'description')) is None:
2824 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2825 if sub_files
is None:
2827 files_to_move
.update(dict(sub_files
))
2829 thumb_files
= self
._write
_thumbnails
(
2830 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2831 if thumb_files
is None:
2833 files_to_move
.update(dict(thumb_files
))
2835 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2836 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2837 if _infojson_written
:
2838 info_dict
['infojson_filename'] = infofn
2839 # For backward compatibility, even though it was a private field
2840 info_dict
['__infojson_filename'] = infofn
2841 elif _infojson_written
is None:
2844 # Note: Annotations are deprecated
2846 if self
.params
.get('writeannotations', False):
2847 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2849 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2851 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2852 self
.to_screen('[info] Video annotations are already present')
2853 elif not info_dict
.get('annotations'):
2854 self
.report_warning('There are no annotations to write.')
2857 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2858 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2859 annofile
.write(info_dict
['annotations'])
2860 except (KeyError, TypeError):
2861 self
.report_warning('There are no annotations to write.')
2862 except (OSError, IOError):
2863 self
.report_error('Cannot write annotations file: ' + annofn
)
2866 # Write internet shortcut files
2867 def _write_link_file(link_type
):
2868 if 'webpage_url' not in info_dict
:
2869 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2871 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2872 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
2874 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2875 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2878 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2879 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2880 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2881 template_vars
= {'url': iri_to_uri(info_dict['webpage_url'])}
2882 if link_type
== 'desktop':
2883 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2884 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
2885 except (OSError, IOError):
2886 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
2891 'url': self
.params
.get('writeurllink'),
2892 'webloc': self
.params
.get('writewebloclink'),
2893 'desktop': self
.params
.get('writedesktoplink'),
2895 if self
.params
.get('writelink'):
2896 link_type
= ('webloc' if sys
.platform
== 'darwin'
2897 else 'desktop' if sys
.platform
.startswith('linux')
2899 write_links
[link_type
] = True
2901 if any(should_write
and not _write_link_file(link_type
)
2902 for link_type
, should_write
in write_links
.items()):
2905 def replace_info_dict(new_info
):
2907 if new_info
== info_dict
:
2910 info_dict
.update(new_info
)
2913 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2914 replace_info_dict(new_info
)
2915 except PostProcessingError
as err
:
2916 self
.report_error('Preprocessing: %s' % str(err
))
2919 if self
.params
.get('skip_download'):
2920 info_dict
['filepath'] = temp_filename
2921 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2922 info_dict
['__files_to_move'] = files_to_move
2923 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
2924 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2927 info_dict
.setdefault('__postprocessors', [])
2930 def existing_video_file(*filepaths
):
2931 ext
= info_dict
.get('ext')
2932 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
2933 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
2934 default_overwrite
=False)
2936 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
2940 if info_dict
.get('requested_formats') is not None:
2942 def compatible_formats(formats
):
2943 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2944 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2945 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2946 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2950 exts
= set(format
.get('ext') for format
in formats
)
2952 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2955 for ext_sets
in COMPATIBLE_EXTS
:
2956 if ext_sets
.issuperset(exts
):
2958 # TODO: Check acodec/vcodec
2961 requested_formats
= info_dict
['requested_formats']
2962 old_ext
= info_dict
['ext']
2963 if self
.params
.get('merge_output_format') is None:
2964 if not compatible_formats(requested_formats
):
2965 info_dict
['ext'] = 'mkv'
2966 self
.report_warning(
2967 'Requested formats are incompatible for merge and will be merged into mkv')
2968 if (info_dict
['ext'] == 'webm'
2969 and info_dict
.get('thumbnails')
2970 # check with type instead of pp_key, __name__, or isinstance
2971 # since we dont want any custom PPs to trigger this
2972 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])):
2973 info_dict
['ext'] = 'mkv'
2974 self
.report_warning(
2975 'webm doesn\'t support embedding a thumbnail, mkv will be used')
2976 new_ext
= info_dict
['ext']
2978 def correct_ext(filename
, ext
=new_ext
):
2981 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2983 os
.path
.splitext(filename
)[0]
2984 if filename_real_ext
in (old_ext
, new_ext
)
2986 return '%s.%s' % (filename_wo_ext
, ext
)
2988 # Ensure filename always has a correct extension for successful merge
2989 full_filename
= correct_ext(full_filename
)
2990 temp_filename
= correct_ext(temp_filename
)
2991 dl_filename
= existing_video_file(full_filename
, temp_filename
)
2992 info_dict
['__real_download'] = False
2995 merger
= FFmpegMergerPP(self
)
2997 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
2998 if dl_filename
is not None:
2999 self
.report_file_already_downloaded(dl_filename
)
3001 for f
in requested_formats
if fd
!= FFmpegFD
else []:
3002 f
['filepath'] = fname
= prepend_extension(
3003 correct_ext(temp_filename
, info_dict
['ext']),
3004 'f%s' % f
['format_id'], info_dict
['ext'])
3005 downloaded
.append(fname
)
3006 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
3007 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3008 info_dict
['__real_download'] = real_download
3010 if self
.params
.get('allow_unplayable_formats'):
3011 self
.report_warning(
3012 'You have requested merging of multiple formats '
3013 'while also allowing unplayable formats to be downloaded. '
3014 'The formats won\'t be merged to prevent data corruption.')
3015 elif not merger
.available
:
3016 self
.report_warning(
3017 'You have requested merging of multiple formats but ffmpeg is not installed. '
3018 'The formats won\'t be merged.')
3020 if temp_filename
== '-':
3021 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3022 else 'but the formats are incompatible for simultaneous download' if merger
.available
3023 else 'but ffmpeg is not installed')
3024 self
.report_warning(
3025 f
'You have requested downloading multiple formats to stdout {reason}. '
3026 'The formats will be streamed one after the other')
3027 fname
= temp_filename
3028 for f
in requested_formats
:
3029 new_info
= dict(info_dict
)
3030 del new_info
['requested_formats']
3032 if temp_filename
!= '-':
3033 fname
= prepend_extension(
3034 correct_ext(temp_filename
, new_info
['ext']),
3035 'f%s' % f
['format_id'], new_info
['ext'])
3036 if not self
._ensure
_dir
_exists
(fname
):
3038 f
['filepath'] = fname
3039 downloaded
.append(fname
)
3040 partial_success
, real_download
= self
.dl(fname
, new_info
)
3041 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3042 success
= success
and partial_success
3044 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3045 info_dict
['__postprocessors'].append(merger
)
3046 info_dict
['__files_to_merge'] = downloaded
3047 # Even if there were no downloads, it is being merged only now
3048 info_dict
['__real_download'] = True
3050 for file in downloaded
:
3051 files_to_move
[file] = None
3053 # Just a single file
3054 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3055 if dl_filename
is None or dl_filename
== temp_filename
:
3056 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3057 # So we should try to resume the download
3058 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3059 info_dict
['__real_download'] = real_download
3061 self
.report_file_already_downloaded(dl_filename
)
3063 dl_filename
= dl_filename
or temp_filename
3064 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3066 except network_exceptions
as err
:
3067 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3069 except (OSError, IOError) as err
:
3070 raise UnavailableVideoError(err
)
3071 except (ContentTooShortError
, ) as err
:
3072 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
3075 if success
and full_filename
!= '-':
3079 fixup_policy
= self
.params
.get('fixup')
3080 vid
= info_dict
['id']
3082 if fixup_policy
in ('ignore', 'never'):
3084 elif fixup_policy
== 'warn':
3086 elif fixup_policy
!= 'force':
3087 assert fixup_policy
in ('detect_or_warn', None)
3088 if not info_dict
.get('__real_download'):
3091 def ffmpeg_fixup(cndn
, msg
, cls
):
3095 self
.report_warning(f
'{vid}: {msg}')
3099 info_dict
['__postprocessors'].append(pp
)
3101 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3103 stretched_ratio
= info_dict
.get('stretched_ratio')
3105 stretched_ratio
not in (1, None),
3106 f
'Non-uniform pixel ratio {stretched_ratio}',
3107 FFmpegFixupStretchedPP
)
3110 (info_dict
.get('requested_formats') is None
3111 and info_dict
.get('container') == 'm4a_dash'
3112 and info_dict
.get('ext') == 'm4a'),
3113 'writing DASH m4a. Only some players support this container',
3116 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3117 downloader
= downloader
.__name
__ if downloader
else None
3119 if info_dict
.get('requested_formats') is None: # Not necessary if doing merger
3120 ffmpeg_fixup(downloader
== 'HlsFD',
3121 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3123 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3124 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3126 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3127 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP
)
3131 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3132 except PostProcessingError
as err
:
3133 self
.report_error('Postprocessing: %s' % str(err
))
3136 for ph
in self
._post
_hooks
:
3137 ph(info_dict
['filepath'])
3138 except Exception as err
:
3139 self
.report_error('post hooks: %s' % str(err
))
3141 info_dict
['__write_download_archive'] = True
3143 if self
.params
.get('force_write_download_archive'):
3144 info_dict
['__write_download_archive'] = True
3146 # Make sure the info_dict was modified in-place
3147 assert info_dict
is original_infodict
3149 max_downloads
= self
.params
.get('max_downloads')
3150 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
3151 raise MaxDownloadsReached()
3153 def __download_wrapper(self
, func
):
3154 @functools.wraps(func
)
3155 def wrapper(*args
, **kwargs
):
3157 res
= func(*args
, **kwargs
)
3158 except UnavailableVideoError
as e
:
3159 self
.report_error(e
)
3160 except MaxDownloadsReached
as e
:
3161 self
.to_screen(f
'[info] {e}')
3163 except DownloadCancelled
as e
:
3164 self
.to_screen(f
'[info] {e}')
3165 if not self
.params
.get('break_per_url'):
3168 if self
.params
.get('dump_single_json', False):
3169 self
.post_extract(res
)
3170 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3173 def download(self
, url_list
):
3174 """Download a given list of URLs."""
3175 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3176 outtmpl
= self
.outtmpl_dict
['default']
3177 if (len(url_list
) > 1
3179 and '%' not in outtmpl
3180 and self
.params
.get('max_downloads') != 1):
3181 raise SameFileError(outtmpl
)
3183 for url
in url_list
:
3184 self
.__download
_wrapper
(self
.extract_info
)(
3185 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3187 return self
._download
_retcode
3189 def download_with_info_file(self
, info_filename
):
3190 with contextlib
.closing(fileinput
.FileInput(
3191 [info_filename
], mode
='r',
3192 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3193 # FileInput doesn't have a read method, we can't call json.load
3194 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3196 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3197 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3198 if not isinstance(e
, EntryNotInPlaylist
):
3199 self
.to_stderr('\r')
3200 webpage_url
= info
.get('webpage_url')
3201 if webpage_url
is not None:
3202 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3203 return self
.download([webpage_url
])
3206 return self
._download
_retcode
3209 def sanitize_info(info_dict
, remove_private_keys
=False):
3210 ''' Sanitize the infodict for converting to json '''
3211 if info_dict
is None:
3213 info_dict
.setdefault('epoch', int(time
.time()))
3214 info_dict
.setdefault('_type', 'video')
3215 remove_keys
= {'__original_infodict'}
# Always remove this since this may contain a copy of the entire dict
3216 keep_keys
= ['_type'] # Always keep this to facilitate load-info-json
3217 if remove_private_keys
:
3219 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3220 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3222 reject
= lambda k
, v
: k
not in keep_keys
and (
3223 k
.startswith('_') or k
in remove_keys
or v
is None)
3225 reject
= lambda k
, v
: k
in remove_keys
3228 if isinstance(obj
, dict):
3229 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3230 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3231 return list(map(filter_fn
, obj
))
3232 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3237 return filter_fn(info_dict
)
3240 def filter_requested_info(info_dict
, actually_filter
=True):
3241 ''' Alias of sanitize_info for backward compatibility '''
3242 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3245 def post_extract(info_dict
):
3246 def actual_post_extract(info_dict
):
3247 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3248 for video_dict
in info_dict
.get('entries', {}):
3249 actual_post_extract(video_dict
or {})
3252 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
3253 extra
= post_extractor().items()
3254 info_dict
.update(extra
)
3255 info_dict
.pop('__post_extractor', None)
3257 original_infodict
= info_dict
.get('__original_infodict') or {}
3258 original_infodict
.update(extra
)
3259 original_infodict
.pop('__post_extractor', None)
3261 actual_post_extract(info_dict
or {})
3263 def run_pp(self
, pp
, infodict
):
3264 files_to_delete
= []
3265 if '__files_to_move' not in infodict
:
3266 infodict
['__files_to_move'] = {}
3268 files_to_delete
, infodict
= pp
.run(infodict
)
3269 except PostProcessingError
as e
:
3270 # Must be True and not 'only_download'
3271 if self
.params
.get('ignoreerrors') is True:
3272 self
.report_error(e
)
3276 if not files_to_delete
:
3278 if self
.params
.get('keepvideo', False):
3279 for f
in files_to_delete
:
3280 infodict
['__files_to_move'].setdefault(f
, '')
3282 for old_filename
in set(files_to_delete
):
3283 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
3285 os
.remove(encodeFilename(old_filename
))
3286 except (IOError, OSError):
3287 self
.report_warning('Unable to remove downloaded original file')
3288 if old_filename
in infodict
['__files_to_move']:
3289 del infodict
['__files_to_move'][old_filename
]
3292 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3293 for tmpl
in self
.params
['forceprint'].get(key
, []):
3294 self
._forceprint
(tmpl
, info
)
3295 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3296 info
= self
.run_pp(pp
, info
)
3299 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3300 info
= dict(ie_info
)
3301 info
['__files_to_move'] = files_to_move
or {}
3302 info
= self
.run_all_pps(key
, info
)
3303 return info
, info
.pop('__files_to_move', None)
3305 def post_process(self
, filename
, info
, files_to_move
=None):
3306 """Run all the postprocessors on the given file."""
3307 info
['filepath'] = filename
3308 info
['__files_to_move'] = files_to_move
or {}
3309 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3310 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3311 del info
['__files_to_move']
3312 return self
.run_all_pps('after_move', info
)
3314 def _make_archive_id(self
, info_dict
):
3315 video_id
= info_dict
.get('id')
3318 # Future-proof against any change in case
3319 # and backwards compatibility with prior versions
3320 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3321 if extractor
is None:
3322 url
= str_or_none(info_dict
.get('url'))
3325 # Try to find matching extractor for the URL and take its ie_key
3326 for ie_key
, ie
in self
._ies
.items():
3327 if ie
.suitable(url
):
3332 return '%s %s' % (extractor
.lower(), video_id
)
3334 def in_download_archive(self
, info_dict
):
3335 fn
= self
.params
.get('download_archive')
3339 vid_id
= self
._make
_archive
_id
(info_dict
)
3341 return False # Incomplete video information
3343 return vid_id
in self
.archive
3345 def record_download_archive(self
, info_dict
):
3346 fn
= self
.params
.get('download_archive')
3349 vid_id
= self
._make
_archive
_id
(info_dict
)
3351 self
.write_debug(f
'Adding to archive: {vid_id}')
3352 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3353 archive_file
.write(vid_id
+ '\n')
3354 self
.archive
.add(vid_id
)
3357 def format_resolution(format
, default
='unknown'):
3358 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3360 if format
.get('resolution') is not None:
3361 return format
['resolution']
3362 if format
.get('width') and format
.get('height'):
3363 return '%dx%d' % (format
['width'], format
['height'])
3364 elif format
.get('height'):
3365 return '%sp' % format
['height']
3366 elif format
.get('width'):
3367 return '%dx?' % format
['width']
3370 def _list_format_headers(self
, *headers
):
3371 if self
.params
.get('listformats_table', True) is not False:
3372 return [self
._format
_screen
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3375 def _format_note(self
, fdict
):
3377 if fdict
.get('ext') in ['f4f', 'f4m']:
3378 res
+= '(unsupported)'
3379 if fdict
.get('language'):
3382 res
+= '[%s]' % fdict
['language']
3383 if fdict
.get('format_note') is not None:
3386 res
+= fdict
['format_note']
3387 if fdict
.get('tbr') is not None:
3390 res
+= '%4dk' % fdict
['tbr']
3391 if fdict
.get('container') is not None:
3394 res
+= '%s container' % fdict
['container']
3395 if (fdict
.get('vcodec') is not None
3396 and fdict
.get('vcodec') != 'none'):
3399 res
+= fdict
['vcodec']
3400 if fdict
.get('vbr') is not None:
3402 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3404 if fdict
.get('vbr') is not None:
3405 res
+= '%4dk' % fdict
['vbr']
3406 if fdict
.get('fps') is not None:
3409 res
+= '%sfps' % fdict
['fps']
3410 if fdict
.get('acodec') is not None:
3413 if fdict
['acodec'] == 'none':
3416 res
+= '%-5s' % fdict
['acodec']
3417 elif fdict
.get('abr') is not None:
3421 if fdict
.get('abr') is not None:
3422 res
+= '@%3dk' % fdict
['abr']
3423 if fdict
.get('asr') is not None:
3424 res
+= ' (%5dHz)' % fdict
['asr']
3425 if fdict
.get('filesize') is not None:
3428 res
+= format_bytes(fdict
['filesize'])
3429 elif fdict
.get('filesize_approx') is not None:
3432 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3435 def render_formats_table(self
, info_dict
):
3436 if not info_dict
.get('formats') and not info_dict
.get('url'):
3439 formats
= info_dict
.get('formats', [info_dict
])
3440 if not self
.params
.get('listformats_table', True) is not False:
3443 format_field(f
, 'format_id'),
3444 format_field(f
, 'ext'),
3445 self
.format_resolution(f
),
3446 self
._format
_note
(f
)
3447 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3448 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3450 delim
= self
._format
_screen
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3453 self
._format
_screen
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3454 format_field(f
, 'ext'),
3455 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3456 format_field(f
, 'fps', '\t%d'),
3457 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3459 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3460 format_field(f
, 'tbr', '\t%dk'),
3461 shorten_protocol_name(f
.get('protocol', '')),
3463 format_field(f
, 'vcodec', default
='unknown').replace(
3464 'none', 'images' if f
.get('acodec') == 'none'
3465 else self
._format
_screen
('audio only', self
.Styles
.SUPPRESS
)),
3466 format_field(f
, 'vbr', '\t%dk'),
3467 format_field(f
, 'acodec', default
='unknown').replace(
3468 'none', '' if f
.get('vcodec') == 'none'
3469 else self
._format
_screen
('video only', self
.Styles
.SUPPRESS
)),
3470 format_field(f
, 'abr', '\t%dk'),
3471 format_field(f
, 'asr', '\t%dHz'),
3473 self
._format
_screen
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3474 format_field(f
, 'language', '[%s]'),
3475 join_nonempty(format_field(f
, 'format_note'),
3476 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3479 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3480 header_line
= self
._list
_format
_headers
(
3481 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3482 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3484 return render_table(
3485 header_line
, table
, hide_empty
=True,
3486 delim
=self
._format
_screen
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3488 def render_thumbnails_table(self
, info_dict
):
3489 thumbnails
= list(info_dict
.get('thumbnails'))
3492 return render_table(
3493 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3494 [[t
.get('id'), t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3496 def render_subtitles_table(self
, video_id
, subtitles
):
3497 def _row(lang
, formats
):
3498 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3499 if len(set(names
)) == 1:
3500 names
= [] if names
[0] == 'unknown' else names
[:1]
3501 return [lang
, ', '.join(names
), ', '.join(exts
)]
3505 return render_table(
3506 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3507 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3510 def __list_table(self
, video_id
, name
, func
, *args
):
3513 self
.to_screen(f
'{video_id} has no {name}')
3515 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3516 self
.to_stdout(table
)
3518 def list_formats(self
, info_dict
):
3519 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3521 def list_thumbnails(self
, info_dict
):
3522 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3524 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3525 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3527 def urlopen(self
, req
):
3528 """ Start an HTTP download """
3529 if isinstance(req
, compat_basestring
):
3530 req
= sanitized_Request(req
)
3531 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3533 def print_debug_header(self
):
3534 if not self
.params
.get('verbose'):
3537 def get_encoding(stream
):
3538 ret
= getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__)
3539 if not supports_terminal_sequences(stream
):
3540 from .compat
import WINDOWS_VT_MODE
3541 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3544 encoding_str
= 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3545 locale
.getpreferredencoding(),
3546 sys
.getfilesystemencoding(),
3547 get_encoding(self
._screen
_file
), get_encoding(self
._err
_file
),
3548 self
.get_encoding())
3550 logger
= self
.params
.get('logger')
3552 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3553 write_debug(encoding_str
)
3555 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3556 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3558 source
= detect_variant()
3559 write_debug(join_nonempty(
3560 'yt-dlp version', __version__
,
3561 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3562 '' if source
== 'unknown' else f
'({source})',
3564 if not _LAZY_LOADER
:
3565 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3566 write_debug('Lazy loading extractors is forcibly disabled')
3568 write_debug('Lazy loading extractors is disabled')
3569 if plugin_extractors
or plugin_postprocessors
:
3570 write_debug('Plugins: %s' % [
3571 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3572 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3573 if self
.params
.get('compat_opts'):
3574 write_debug('Compatibility options: %s' % ', '.join(self
.params
.get('compat_opts')))
3576 if source
== 'source':
3579 ['git', 'rev-parse', '--short', 'HEAD'],
3580 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3581 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3582 out
, err
= sp
.communicate_or_kill()
3583 out
= out
.decode().strip()
3584 if re
.match('[0-9a-f]+', out
):
3585 write_debug('Git HEAD: %s' % out
)
3592 def python_implementation():
3593 impl_name
= platform
.python_implementation()
3594 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3595 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3598 write_debug('Python version %s (%s %s) - %s' % (
3599 platform
.python_version(),
3600 python_implementation(),
3601 platform
.architecture()[0],
3604 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3605 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3607 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features
)
3609 exe_versions
['rtmpdump'] = rtmpdump_version()
3610 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3611 exe_str
= ', '.join(
3612 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3614 write_debug('exe versions: %s' % exe_str
)
3616 from .downloader
.websocket
import has_websockets
3617 from .postprocessor
.embedthumbnail
import has_mutagen
3618 from .cookies
import SQLITE_AVAILABLE
, SECRETSTORAGE_AVAILABLE
3620 lib_str
= join_nonempty(
3621 compat_pycrypto_AES
and compat_pycrypto_AES
.__name
__.split('.')[0],
3622 SECRETSTORAGE_AVAILABLE
and 'secretstorage',
3623 has_mutagen
and 'mutagen',
3624 SQLITE_AVAILABLE
and 'sqlite',
3625 has_websockets
and 'websockets',
3626 delim
=', ') or 'none'
3627 write_debug('Optional libraries: %s' % lib_str
)
3630 for handler
in self
._opener
.handlers
:
3631 if hasattr(handler
, 'proxies'):
3632 proxy_map
.update(handler
.proxies
)
3633 write_debug(f
'Proxy map: {proxy_map}')
3636 if False and self
.params
.get('call_home'):
3637 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3638 write_debug('Public IP address: %s' % ipaddr
)
3639 latest_version
= self
.urlopen(
3640 'https://yt-dl.org/latest/version').read().decode('utf-8')
3641 if version_tuple(latest_version
) > version_tuple(__version__
):
3642 self
.report_warning(
3643 'You are using an outdated version (newest version: %s)! '
3644 'See https://yt-dl.org/update if you need help updating.' %
3647 def _setup_opener(self
):
3648 timeout_val
= self
.params
.get('socket_timeout')
3649 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3651 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3652 opts_cookiefile
= self
.params
.get('cookiefile')
3653 opts_proxy
= self
.params
.get('proxy')
3655 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3657 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3658 if opts_proxy
is not None:
3659 if opts_proxy
== '':
3662 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3664 proxies
= compat_urllib_request
.getproxies()
3665 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3666 if 'http' in proxies
and 'https' not in proxies
:
3667 proxies
['https'] = proxies
['http']
3668 proxy_handler
= PerRequestProxyHandler(proxies
)
3670 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3671 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3672 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3673 redirect_handler
= YoutubeDLRedirectHandler()
3674 data_handler
= compat_urllib_request_DataHandler()
3676 # When passing our own FileHandler instance, build_opener won't add the
3677 # default FileHandler and allows us to disable the file protocol, which
3678 # can be used for malicious purposes (see
3679 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3680 file_handler
= compat_urllib_request
.FileHandler()
3682 def file_open(*args
, **kwargs
):
3683 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3684 file_handler
.file_open
= file_open
3686 opener
= compat_urllib_request
.build_opener(
3687 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3689 # Delete the default user-agent header, which would otherwise apply in
3690 # cases where our custom HTTP handler doesn't come into play
3691 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3692 opener
.addheaders
= []
3693 self
._opener
= opener
3695 def encode(self
, s
):
3696 if isinstance(s
, bytes):
3697 return s
# Already encoded
3700 return s
.encode(self
.get_encoding())
3701 except UnicodeEncodeError as err
:
3702 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3705 def get_encoding(self
):
3706 encoding
= self
.params
.get('encoding')
3707 if encoding
is None:
3708 encoding
= preferredencoding()
3711 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3712 ''' Write infojson and returns True = written, False = skip, None = error '''
3713 if overwrite
is None:
3714 overwrite
= self
.params
.get('overwrites', True)
3715 if not self
.params
.get('writeinfojson'):
3718 self
.write_debug(f
'Skipping writing {label} infojson')
3720 elif not self
._ensure
_dir
_exists
(infofn
):
3722 elif not overwrite
and os
.path
.exists(infofn
):
3723 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3725 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3727 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3728 except (OSError, IOError):
3729 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3733 def _write_description(self
, label
, ie_result
, descfn
):
3734 ''' Write description and returns True = written, False = skip, None = error '''
3735 if not self
.params
.get('writedescription'):
3738 self
.write_debug(f
'Skipping writing {label} description')
3740 elif not self
._ensure
_dir
_exists
(descfn
):
3742 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3743 self
.to_screen(f
'[info] {label.title()} description is already present')
3744 elif ie_result
.get('description') is None:
3745 self
.report_warning(f
'There\'s no {label} description to write')
3749 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3750 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3751 descfile
.write(ie_result
['description'])
3752 except (OSError, IOError):
3753 self
.report_error(f
'Cannot write {label} description file {descfn}')
3757 def _write_subtitles(self
, info_dict
, filename
):
3758 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3760 subtitles
= info_dict
.get('requested_subtitles')
3761 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3762 # subtitles download errors are already managed as troubles in relevant IE
3763 # that way it will silently go on when used with unsupporting IE
3766 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3767 if not sub_filename_base
:
3768 self
.to_screen('[info] Skipping writing video subtitles')
3770 for sub_lang
, sub_info
in subtitles
.items():
3771 sub_format
= sub_info
['ext']
3772 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3773 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3774 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3776 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3777 sub_info
['filepath'] = existing_sub
3778 ret
.append((existing_sub
, sub_filename_final
))
3781 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3782 if sub_info
.get('data') is not None:
3784 # Use newline='' to prevent conversion of newline characters
3785 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3786 with io
.open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3787 subfile
.write(sub_info
['data'])
3788 sub_info
['filepath'] = sub_filename
3789 ret
.append((sub_filename
, sub_filename_final
))
3791 except (OSError, IOError):
3792 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3796 sub_copy
= sub_info
.copy()
3797 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3798 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3799 sub_info
['filepath'] = sub_filename
3800 ret
.append((sub_filename
, sub_filename_final
))
3801 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3802 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3803 raise DownloadError(f
'Unable to download video subtitles for {sub_lang!r}: {err}', err
)
3804 self
.report_warning(f
'Unable to download video subtitles for {sub_lang!r}: {err}')
3807 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3808 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3809 write_all
= self
.params
.get('write_all_thumbnails', False)
3810 thumbnails
, ret
= [], []
3811 if write_all
or self
.params
.get('writethumbnail', False):
3812 thumbnails
= info_dict
.get('thumbnails') or []
3813 multiple
= write_all
and len(thumbnails
) > 1
3815 if thumb_filename_base
is None:
3816 thumb_filename_base
= filename
3817 if thumbnails
and not thumb_filename_base
:
3818 self
.write_debug(f
'Skipping writing {label} thumbnail')
3821 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3822 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3823 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3824 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3825 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3827 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3829 self
.to_screen('[info] %s is already present' % (
3830 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3831 t
['filepath'] = existing_thumb
3832 ret
.append((existing_thumb
, thumb_filename_final
))
3834 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3836 uf
= self
.urlopen(t
['url'])
3837 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3838 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3839 shutil
.copyfileobj(uf
, thumbf
)
3840 ret
.append((thumb_filename
, thumb_filename_final
))
3841 t
['filepath'] = thumb_filename
3842 except network_exceptions
as err
:
3844 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3845 if ret
and not write_all
: