4 from __future__
import absolute_import
, unicode_literals
31 from string
import ascii_letters
35 compat_get_terminal_size
,
42 compat_tokenize_tokenize
,
44 compat_urllib_request
,
45 compat_urllib_request_DataHandler
,
46 windows_enable_vt_mode
,
48 from .cookies
import load_cookies
70 format_decimal_suffix
,
91 PerRequestProxyHandler
,
99 register_socks_protocols
,
100 RejectedVideoReached
,
101 remove_terminal_sequences
,
115 supports_terminal_sequences
,
120 UnavailableVideoError
,
126 YoutubeDLCookieProcessor
,
128 YoutubeDLRedirectHandler
,
130 from .cache
import Cache
131 from .minicurses
import format_text
132 from .extractor
import (
133 gen_extractor_classes
,
136 _PLUGIN_CLASSES
as plugin_extractors
138 from .extractor
.openload
import PhantomJSwrapper
139 from .downloader
import (
141 get_suitable_downloader
,
142 shorten_protocol_name
144 from .downloader
.rtmp
import rtmpdump_version
145 from .postprocessor
import (
148 FFmpegFixupDuplicateMoovPP
,
149 FFmpegFixupDurationPP
,
152 FFmpegFixupStretchedPP
,
153 FFmpegFixupTimestampPP
,
156 MoveFilesAfterDownloadPP
,
157 _PLUGIN_CLASSES
as plugin_postprocessors
159 from .update
import detect_variant
160 from .version
import __version__
, RELEASE_GIT_HEAD
162 if compat_os_name
== 'nt':
166 class YoutubeDL(object):
169 YoutubeDL objects are the ones responsible of downloading the
170 actual video file and writing it to disk if the user has requested
171 it, among some other tasks. In most cases there should be one per
172 program. As, given a video URL, the downloader doesn't know how to
173 extract all the needed information, task that InfoExtractors do, it
174 has to pass the URL to one of them.
176 For this, YoutubeDL objects have a method that allows
177 InfoExtractors to be registered in a given order. When it is passed
178 a URL, the YoutubeDL object handles it to the first InfoExtractor it
179 finds that reports being able to handle it. The InfoExtractor extracts
180 all the information about the video or videos the URL refers to, and
181 YoutubeDL process the extracted information, possibly using a File
182 Downloader to download the video.
184 YoutubeDL objects accept a lot of parameters. In order not to saturate
185 the object constructor with arguments, it receives a dictionary of
186 options instead. These options are available through the params
187 attribute for the InfoExtractors to use. The YoutubeDL also
188 registers itself as the downloader in charge for the InfoExtractors
189 that are added to it, so this is a "mutual registration".
193 username: Username for authentication purposes.
194 password: Password for authentication purposes.
195 videopassword: Password for accessing a video.
196 ap_mso: Adobe Pass multiple-system operator identifier.
197 ap_username: Multiple-system operator account username.
198 ap_password: Multiple-system operator account password.
199 usenetrc: Use netrc for authentication instead.
200 verbose: Print additional info to stdout.
201 quiet: Do not print messages to stdout.
202 no_warnings: Do not print out anything for warnings.
203 forceprint: A dict with keys video/playlist mapped to
204 a list of templates to force print to stdout
205 For compatibility, a single list is also accepted
206 forceurl: Force printing final URL. (Deprecated)
207 forcetitle: Force printing title. (Deprecated)
208 forceid: Force printing ID. (Deprecated)
209 forcethumbnail: Force printing thumbnail URL. (Deprecated)
210 forcedescription: Force printing description. (Deprecated)
211 forcefilename: Force printing final filename. (Deprecated)
212 forceduration: Force printing duration. (Deprecated)
213 forcejson: Force printing info_dict as JSON.
214 dump_single_json: Force printing the info_dict of the whole playlist
215 (or video) as a single JSON line.
216 force_write_download_archive: Force writing download archive regardless
217 of 'skip_download' or 'simulate'.
218 simulate: Do not download the video files. If unset (or None),
219 simulate only if listsubtitles, listformats or list_thumbnails is used
220 format: Video format code. see "FORMAT SELECTION" for more details.
221 You can also pass a function. The function takes 'ctx' as
222 argument and returns the formats to download.
223 See "build_format_selector" for an implementation
224 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
225 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
226 extracting metadata even if the video is not actually
227 available for download (experimental)
228 format_sort: A list of fields by which to sort the video formats.
229 See "Sorting Formats" for more details.
230 format_sort_force: Force the given format_sort. see "Sorting Formats"
232 allow_multiple_video_streams: Allow multiple video streams to be merged
234 allow_multiple_audio_streams: Allow multiple audio streams to be merged
236 check_formats Whether to test if the formats are downloadable.
237 Can be True (check all), False (check none),
238 'selected' (check selected formats),
239 or None (check only if requested by extractor)
240 paths: Dictionary of output paths. The allowed keys are 'home'
241 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
242 outtmpl: Dictionary of templates for output names. Allowed keys
243 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
244 For compatibility with youtube-dl, a single string can also be used
245 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
246 restrictfilenames: Do not allow "&" and spaces in file names
247 trim_file_name: Limit length of filename (extension excluded)
248 windowsfilenames: Force the filenames to be windows compatible
249 ignoreerrors: Do not stop on download/postprocessing errors.
250 Can be 'only_download' to ignore only download errors.
251 Default is 'only_download' for CLI, but False for API
252 skip_playlist_after_errors: Number of allowed failures until the rest of
253 the playlist is skipped
254 force_generic_extractor: Force downloader to use the generic extractor
255 overwrites: Overwrite all video and metadata files if True,
256 overwrite only non-video files if None
257 and don't overwrite any file if False
258 For compatibility with youtube-dl,
259 "nooverwrites" may also be used instead
260 playliststart: Playlist item to start at.
261 playlistend: Playlist item to end at.
262 playlist_items: Specific indices of playlist to download.
263 playlistreverse: Download playlist items in reverse order.
264 playlistrandom: Download playlist items in random order.
265 matchtitle: Download only matching titles.
266 rejecttitle: Reject downloads for matching titles.
267 logger: Log messages to a logging.Logger instance.
268 logtostderr: Log messages to stderr instead of stdout.
269 consoletitle: Display progress in console window's titlebar.
270 writedescription: Write the video description to a .description file
271 writeinfojson: Write the video description to a .info.json file
272 clean_infojson: Remove private fields from the infojson
273 getcomments: Extract video comments. This will not be written to disk
274 unless writeinfojson is also given
275 writeannotations: Write the video annotations to a .annotations.xml file
276 writethumbnail: Write the thumbnail image to a file
277 allow_playlist_files: Whether to write playlists' description, infojson etc
278 also to disk when using the 'write*' options
279 write_all_thumbnails: Write all thumbnail formats to files
280 writelink: Write an internet shortcut file, depending on the
281 current platform (.url/.webloc/.desktop)
282 writeurllink: Write a Windows internet shortcut file (.url)
283 writewebloclink: Write a macOS internet shortcut file (.webloc)
284 writedesktoplink: Write a Linux internet shortcut file (.desktop)
285 writesubtitles: Write the video subtitles to a file
286 writeautomaticsub: Write the automatically generated subtitles to a file
287 allsubtitles: Deprecated - Use subtitleslangs = ['all']
288 Downloads all the subtitles of the video
289 (requires writesubtitles or writeautomaticsub)
290 listsubtitles: Lists all available subtitles for the video
291 subtitlesformat: The format code for subtitles
292 subtitleslangs: List of languages of the subtitles to download (can be regex).
293 The list may contain "all" to refer to all the available
294 subtitles. The language can be prefixed with a "-" to
295 exclude it from the requested languages. Eg: ['all', '-live_chat']
296 keepvideo: Keep the video file after post-processing
297 daterange: A DateRange object, download only if the upload_date is in the range.
298 skip_download: Skip the actual download of the video file
299 cachedir: Location of the cache files in the filesystem.
300 False to disable filesystem cache.
301 noplaylist: Download single video instead of a playlist if in doubt.
302 age_limit: An integer representing the user's age in years.
303 Unsuitable videos for the given age are skipped.
304 min_views: An integer representing the minimum view count the video
305 must have in order to not be skipped.
306 Videos without view count information are always
307 downloaded. None for no limit.
308 max_views: An integer representing the maximum view count.
309 Videos that are more popular than that are not
311 Videos without view count information are always
312 downloaded. None for no limit.
313 download_archive: File name of a file where all downloads are recorded.
314 Videos already present in the file are not downloaded
316 break_on_existing: Stop the download process after attempting to download a
317 file that is in the archive.
318 break_on_reject: Stop the download process when encountering a video that
319 has been filtered out.
320 break_per_url: Whether break_on_reject and break_on_existing
321 should act on each input URL as opposed to for the entire queue
322 cookiefile: File name where cookies should be read from and dumped to
323 cookiesfrombrowser: A tuple containing the name of the browser, the profile
324 name/pathfrom where cookies are loaded, and the name of the
325 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
326 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
327 support RFC 5746 secure renegotiation
328 nocheckcertificate: Do not verify SSL certificates
329 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
330 At the moment, this is only supported by YouTube.
331 proxy: URL of the proxy server to use
332 geo_verification_proxy: URL of the proxy to use for IP address verification
333 on geo-restricted sites.
334 socket_timeout: Time to wait for unresponsive hosts, in seconds
335 bidi_workaround: Work around buggy terminals without bidirectional text
336 support, using fridibi
337 debug_printtraffic:Print out sent and received HTTP traffic
338 include_ads: Download ads as well (deprecated)
339 default_search: Prepend this string if an input url is not valid.
340 'auto' for elaborate guessing
341 encoding: Use this encoding instead of the system-specified.
342 extract_flat: Do not resolve URLs, return the immediate result.
343 Pass in 'in_playlist' to only show this behavior for
345 wait_for_video: If given, wait for scheduled streams to become available.
346 The value should be a tuple containing the range
347 (min_secs, max_secs) to wait between retries
348 postprocessors: A list of dictionaries, each with an entry
349 * key: The name of the postprocessor. See
350 yt_dlp/postprocessor/__init__.py for a list.
351 * when: When to run the postprocessor. Can be one of
352 pre_process|before_dl|post_process|after_move.
353 Assumed to be 'post_process' if not given
354 post_hooks: Deprecated - Register a custom postprocessor instead
355 A list of functions that get called as the final step
356 for each video file, after all postprocessors have been
357 called. The filename will be passed as the only argument.
358 progress_hooks: A list of functions that get called on download
359 progress, with a dictionary with the entries
360 * status: One of "downloading", "error", or "finished".
361 Check this first and ignore unknown values.
362 * info_dict: The extracted info_dict
364 If status is one of "downloading", or "finished", the
365 following properties may also be present:
366 * filename: The final filename (always present)
367 * tmpfilename: The filename we're currently writing to
368 * downloaded_bytes: Bytes on disk
369 * total_bytes: Size of the whole file, None if unknown
370 * total_bytes_estimate: Guess of the eventual file size,
372 * elapsed: The number of seconds since download started.
373 * eta: The estimated time in seconds, None if unknown
374 * speed: The download speed in bytes/second, None if
376 * fragment_index: The counter of the currently
377 downloaded video fragment.
378 * fragment_count: The number of fragments (= individual
379 files that will be merged)
381 Progress hooks are guaranteed to be called at least once
382 (with status "finished") if the download is successful.
383 postprocessor_hooks: A list of functions that get called on postprocessing
384 progress, with a dictionary with the entries
385 * status: One of "started", "processing", or "finished".
386 Check this first and ignore unknown values.
387 * postprocessor: Name of the postprocessor
388 * info_dict: The extracted info_dict
390 Progress hooks are guaranteed to be called at least twice
391 (with status "started" and "finished") if the processing is successful.
392 merge_output_format: Extension to use when merging formats.
393 final_ext: Expected final extension; used to detect when the file was
394 already downloaded and converted
395 fixup: Automatically correct known faults of the file.
397 - "never": do nothing
398 - "warn": only emit a warning
399 - "detect_or_warn": check whether we can do anything
400 about it, warn otherwise (default)
401 source_address: Client-side IP address to bind to.
402 call_home: Boolean, true iff we are allowed to contact the
403 yt-dlp servers for debugging. (BROKEN)
404 sleep_interval_requests: Number of seconds to sleep between requests
406 sleep_interval: Number of seconds to sleep before each download when
407 used alone or a lower bound of a range for randomized
408 sleep before each download (minimum possible number
409 of seconds to sleep) when used along with
411 max_sleep_interval:Upper bound of a range for randomized sleep before each
412 download (maximum possible number of seconds to sleep).
413 Must only be used along with sleep_interval.
414 Actual sleep time will be a random float from range
415 [sleep_interval; max_sleep_interval].
416 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
417 listformats: Print an overview of available video formats and exit.
418 list_thumbnails: Print a table of all thumbnails and exit.
419 match_filter: A function that gets called with the info_dict of
421 If it returns a message, the video is ignored.
422 If it returns None, the video is downloaded.
423 match_filter_func in utils.py is one example for this.
424 no_color: Do not emit color codes in output.
425 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
428 Two-letter ISO 3166-2 country code that will be used for
429 explicit geographic restriction bypassing via faking
430 X-Forwarded-For HTTP header
432 IP range in CIDR notation that will be used similarly to
435 The following options determine which downloader is picked:
436 external_downloader: A dictionary of protocol keys and the executable of the
437 external downloader to use for it. The allowed protocols
438 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
439 Set the value to 'native' to use the native downloader
440 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
441 or {'m3u8': 'ffmpeg'} instead.
442 Use the native HLS downloader instead of ffmpeg/avconv
443 if True, otherwise use ffmpeg/avconv if False, otherwise
444 use downloader suggested by extractor if None.
445 compat_opts: Compatibility options. See "Differences in default behavior".
446 The following options do not work when used through the API:
447 filename, abort-on-error, multistreams, no-live-chat, format-sort
448 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
449 Refer __init__.py for their implementation
450 progress_template: Dictionary of templates for progress outputs.
451 Allowed keys are 'download', 'postprocess',
452 'download-title' (console title) and 'postprocess-title'.
453 The template is mapped on a dictionary with keys 'progress' and 'info'
455 The following parameters are not used by YoutubeDL itself, they are used by
456 the downloader (see yt_dlp/downloader/common.py):
457 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
458 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
459 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
460 external_downloader_args, concurrent_fragment_downloads.
462 The following options are used by the post processors:
463 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
464 otherwise prefer ffmpeg. (avconv support is deprecated)
465 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
466 to the binary or its containing directory.
467 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
468 and a list of additional command-line arguments for the
469 postprocessor/executable. The dict can also have "PP+EXE" keys
470 which are used when the given exe is used by the given PP.
471 Use 'default' as the name for arguments to passed to all PP
472 For compatibility with youtube-dl, a single list of args
475 The following options are used by the extractors:
476 extractor_retries: Number of times to retry for known errors
477 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
478 hls_split_discontinuity: Split HLS playlists to different formats at
479 discontinuities such as ad breaks (default: False)
480 extractor_args: A dictionary of arguments to be passed to the extractors.
481 See "EXTRACTOR ARGUMENTS" for details.
482 Eg: {'youtube': {'skip': ['dash', 'hls']}}
483 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
484 If True (default), DASH manifests and related
485 data will be downloaded and processed by extractor.
486 You can reduce network I/O by disabling it if you don't
487 care about DASH. (only for youtube)
488 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
489 If True (default), HLS manifests and related
490 data will be downloaded and processed by extractor.
491 You can reduce network I/O by disabling it if you don't
492 care about HLS. (only for youtube)
495 _NUMERIC_FIELDS
= set((
496 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
497 'timestamp', 'release_timestamp',
498 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
499 'average_rating', 'comment_count', 'age_limit',
500 'start_time', 'end_time',
501 'chapter_number', 'season_number', 'episode_number',
502 'track_number', 'disc_number', 'release_year',
505 _format_selection_exts
= {
506 'audio': {'m4a', 'mp3', 'ogg', 'aac'}
,
507 'video': {'mp4', 'flv', 'webm', '3gp'}
,
508 'storyboards': {'mhtml'}
,
513 _pps
= {k: [] for k in POSTPROCESS_WHEN}
514 _printed_messages
= set()
515 _first_webpage_request
= True
516 _download_retcode
= None
517 _num_downloads
= None
519 _playlist_urls
= set()
522 def __init__(self
, params
=None, auto_init
=True):
523 """Create a FileDownloader object with the given options.
524 @param auto_init Whether to load the default extractors and print header (if verbose).
525 Set to 'no_verbose_header' to not print the header
530 self
._ies
_instances
= {}
531 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
532 self
._printed
_messages
= set()
533 self
._first
_webpage
_request
= True
534 self
._post
_hooks
= []
535 self
._progress
_hooks
= []
536 self
._postprocessor
_hooks
= []
537 self
._download
_retcode
= 0
538 self
._num
_downloads
= 0
540 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
541 self
._err
_file
= sys
.stderr
543 self
.cache
= Cache(self
)
545 windows_enable_vt_mode()
546 self
._allow
_colors
= {
547 'screen': not self
.params
.get('no_color') and supports_terminal_sequences(self
._screen
_file
),
548 'err': not self
.params
.get('no_color') and supports_terminal_sequences(self
._err
_file
),
551 if sys
.version_info
< (3, 6):
553 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
555 if self
.params
.get('allow_unplayable_formats'):
557 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
558 'This is a developer option intended for debugging. \n'
559 ' If you experience any issues while using this option, '
560 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
562 def check_deprecated(param
, option
, suggestion
):
563 if self
.params
.get(param
) is not None:
564 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
568 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
569 if self
.params
.get('geo_verification_proxy') is None:
570 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
572 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
573 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
574 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
576 for msg
in self
.params
.get('_warnings', []):
577 self
.report_warning(msg
)
578 for msg
in self
.params
.get('_deprecation_warnings', []):
579 self
.deprecation_warning(msg
)
581 if 'list-formats' in self
.params
.get('compat_opts', []):
582 self
.params
['listformats_table'] = False
584 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
585 # nooverwrites was unnecessarily changed to overwrites
586 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
587 # This ensures compatibility with both keys
588 self
.params
['overwrites'] = not self
.params
['nooverwrites']
589 elif self
.params
.get('overwrites') is None:
590 self
.params
.pop('overwrites', None)
592 self
.params
['nooverwrites'] = not self
.params
['overwrites']
594 # Compatibility with older syntax
595 params
.setdefault('forceprint', {})
596 if not isinstance(params
['forceprint'], dict):
597 params
['forceprint'] = {'video': params['forceprint']}
599 if params
.get('bidi_workaround', False):
602 master
, slave
= pty
.openpty()
603 width
= compat_get_terminal_size().columns
607 width_args
= ['-w', str(width
)]
609 stdin
=subprocess
.PIPE
,
611 stderr
=self
._err
_file
)
613 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
615 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
616 self
._output
_channel
= os
.fdopen(master
, 'rb')
617 except OSError as ose
:
618 if ose
.errno
== errno
.ENOENT
:
620 'Could not find fribidi executable, ignoring --bidi-workaround. '
621 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
625 if (sys
.platform
!= 'win32'
626 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
627 and not params
.get('restrictfilenames', False)):
628 # Unicode filesystem API will throw errors (#1474, #13027)
630 'Assuming --restrict-filenames since file system encoding '
631 'cannot encode all characters. '
632 'Set the LC_ALL environment variable to fix this.')
633 self
.params
['restrictfilenames'] = True
635 self
.outtmpl_dict
= self
.parse_outtmpl()
637 # Creating format selector here allows us to catch syntax errors before the extraction
638 self
.format_selector
= (
639 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
640 else self
.params
['format'] if callable(self
.params
['format'])
641 else self
.build_format_selector(self
.params
['format']))
646 if auto_init
!= 'no_verbose_header':
647 self
.print_debug_header()
648 self
.add_default_info_extractors()
651 'post_hooks': self
.add_post_hook
,
652 'progress_hooks': self
.add_progress_hook
,
653 'postprocessor_hooks': self
.add_postprocessor_hook
,
655 for opt
, fn
in hooks
.items():
656 for ph
in self
.params
.get(opt
, []):
659 for pp_def_raw
in self
.params
.get('postprocessors', []):
660 pp_def
= dict(pp_def_raw
)
661 when
= pp_def
.pop('when', 'post_process')
662 self
.add_post_processor(
663 get_postprocessor(pp_def
.pop('key'))(self
, **compat_kwargs(pp_def
)),
666 register_socks_protocols()
668 def preload_download_archive(fn
):
669 """Preload the archive, if any is specified"""
672 self
.write_debug(f
'Loading archive file {fn!r}')
674 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
675 for line
in archive_file
:
676 self
.archive
.add(line
.strip())
677 except IOError as ioe
:
678 if ioe
.errno
!= errno
.ENOENT
:
684 preload_download_archive(self
.params
.get('download_archive'))
686 def warn_if_short_id(self
, argv
):
687 # short YouTube ID starting with dash?
689 i
for i
, a
in enumerate(argv
)
690 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
694 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
695 + ['--'] + [argv
[i
] for i
in idxs
]
698 'Long argument string detected. '
699 'Use -- to separate parameters and URLs, like this:\n%s' %
700 args_to_str(correct_argv
))
702 def add_info_extractor(self
, ie
):
703 """Add an InfoExtractor object to the end of the list."""
705 self
._ies
[ie_key
] = ie
706 if not isinstance(ie
, type):
707 self
._ies
_instances
[ie_key
] = ie
708 ie
.set_downloader(self
)
710 def _get_info_extractor_class(self
, ie_key
):
711 ie
= self
._ies
.get(ie_key
)
713 ie
= get_info_extractor(ie_key
)
714 self
.add_info_extractor(ie
)
717 def get_info_extractor(self
, ie_key
):
719 Get an instance of an IE with name ie_key, it will try to get one from
720 the _ies list, if there's no instance it will create a new one and add
721 it to the extractor list.
723 ie
= self
._ies
_instances
.get(ie_key
)
725 ie
= get_info_extractor(ie_key
)()
726 self
.add_info_extractor(ie
)
729 def add_default_info_extractors(self
):
731 Add the InfoExtractors returned by gen_extractors to the end of the list
733 for ie
in gen_extractor_classes():
734 self
.add_info_extractor(ie
)
736 def add_post_processor(self
, pp
, when
='post_process'):
737 """Add a PostProcessor object to the end of the chain."""
738 self
._pps
[when
].append(pp
)
739 pp
.set_downloader(self
)
741 def add_post_hook(self
, ph
):
742 """Add the post hook"""
743 self
._post
_hooks
.append(ph
)
745 def add_progress_hook(self
, ph
):
746 """Add the download progress hook"""
747 self
._progress
_hooks
.append(ph
)
749 def add_postprocessor_hook(self
, ph
):
750 """Add the postprocessing progress hook"""
751 self
._postprocessor
_hooks
.append(ph
)
752 for pps
in self
._pps
.values():
754 pp
.add_progress_hook(ph
)
756 def _bidi_workaround(self
, message
):
757 if not hasattr(self
, '_output_channel'):
760 assert hasattr(self
, '_output_process')
761 assert isinstance(message
, compat_str
)
762 line_count
= message
.count('\n') + 1
763 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
764 self
._output
_process
.stdin
.flush()
765 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
766 for _
in range(line_count
))
767 return res
[:-len('\n')]
769 def _write_string(self
, message
, out
=None, only_once
=False):
771 if message
in self
._printed
_messages
:
773 self
._printed
_messages
.add(message
)
774 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
776 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
777 """Print message to stdout"""
778 if self
.params
.get('logger'):
779 self
.params
['logger'].debug(message
)
780 elif not quiet
or self
.params
.get('verbose'):
782 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
783 self
._err
_file
if quiet
else self
._screen
_file
)
785 def to_stderr(self
, message
, only_once
=False):
786 """Print message to stderr"""
787 assert isinstance(message
, compat_str
)
788 if self
.params
.get('logger'):
789 self
.params
['logger'].error(message
)
791 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._err
_file
, only_once
=only_once
)
793 def to_console_title(self
, message
):
794 if not self
.params
.get('consoletitle', False):
796 message
= remove_terminal_sequences(message
)
797 if compat_os_name
== 'nt':
798 if ctypes
.windll
.kernel32
.GetConsoleWindow():
799 # c_wchar_p() might not be necessary if `message` is
800 # already of type unicode()
801 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
802 elif 'TERM' in os
.environ
:
803 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
805 def save_console_title(self
):
806 if not self
.params
.get('consoletitle', False):
808 if self
.params
.get('simulate'):
810 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
811 # Save the title on stack
812 self
._write
_string
('\033[22;0t', self
._screen
_file
)
814 def restore_console_title(self
):
815 if not self
.params
.get('consoletitle', False):
817 if self
.params
.get('simulate'):
819 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
820 # Restore the title from stack
821 self
._write
_string
('\033[23;0t', self
._screen
_file
)
824 self
.save_console_title()
827 def __exit__(self
, *args
):
828 self
.restore_console_title()
830 if self
.params
.get('cookiefile') is not None:
831 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
833 def trouble(self
, message
=None, tb
=None, is_error
=True):
834 """Determine action to take when a download problem appears.
836 Depending on if the downloader has been configured to ignore
837 download errors or not, this method may throw an exception or
838 not when errors are found, after printing the message.
840 @param tb If given, is additional traceback information
841 @param is_error Whether to raise error according to ignorerrors
843 if message
is not None:
844 self
.to_stderr(message
)
845 if self
.params
.get('verbose'):
847 if sys
.exc_info()[0]: # if .trouble has been called from an except block
849 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
850 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
851 tb
+= encode_compat_str(traceback
.format_exc())
853 tb_data
= traceback
.format_list(traceback
.extract_stack())
854 tb
= ''.join(tb_data
)
859 if not self
.params
.get('ignoreerrors'):
860 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
861 exc_info
= sys
.exc_info()[1].exc_info
863 exc_info
= sys
.exc_info()
864 raise DownloadError(message
, exc_info
)
865 self
._download
_retcode
= 1
867 def to_screen(self
, message
, skip_eol
=False):
868 """Print message to stdout if not in quiet mode"""
870 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
874 EMPHASIS
= 'light blue'
879 SUPPRESS
= 'light black'
881 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
884 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', 'ascii')
885 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
886 if fallback
is not None and text
!= original_text
:
888 if isinstance(f
, self
.Styles
):
890 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
892 def _format_screen(self
, *args
, **kwargs
):
893 return self
._format
_text
(
894 self
._screen
_file
, self
._allow
_colors
['screen'], *args
, **kwargs
)
896 def _format_err(self
, *args
, **kwargs
):
897 return self
._format
_text
(
898 self
._err
_file
, self
._allow
_colors
['err'], *args
, **kwargs
)
900 def report_warning(self
, message
, only_once
=False):
902 Print the message to stderr, it will be prefixed with 'WARNING:'
903 If stderr is a tty file the 'WARNING:' will be colored
905 if self
.params
.get('logger') is not None:
906 self
.params
['logger'].warning(message
)
908 if self
.params
.get('no_warnings'):
910 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
912 def deprecation_warning(self
, message
):
913 if self
.params
.get('logger') is not None:
914 self
.params
['logger'].warning('DeprecationWarning: {message}')
916 self
.to_stderr(f
'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
918 def report_error(self
, message
, *args
, **kwargs
):
920 Do the same as trouble, but prefixes the message with 'ERROR:', colored
921 in red if stderr is a tty file.
923 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
925 def write_debug(self
, message
, only_once
=False):
926 '''Log debug message or Print message to stderr'''
927 if not self
.params
.get('verbose', False):
929 message
= '[debug] %s' % message
930 if self
.params
.get('logger'):
931 self
.params
['logger'].debug(message
)
933 self
.to_stderr(message
, only_once
)
935 def report_file_already_downloaded(self
, file_name
):
936 """Report file has already been fully downloaded."""
938 self
.to_screen('[download] %s has already been downloaded' % file_name
)
939 except UnicodeEncodeError:
940 self
.to_screen('[download] The file has already been downloaded')
942 def report_file_delete(self
, file_name
):
943 """Report that existing file will be deleted."""
945 self
.to_screen('Deleting existing file %s' % file_name
)
946 except UnicodeEncodeError:
947 self
.to_screen('Deleting existing file')
949 def raise_no_formats(self
, info
, forced
=False):
950 has_drm
= info
.get('__has_drm')
951 msg
= 'This video is DRM protected' if has_drm
else 'No video formats found!'
952 expected
= self
.params
.get('ignore_no_formats_error')
953 if forced
or not expected
:
954 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
955 expected
=has_drm
or expected
)
957 self
.report_warning(msg
)
959 def parse_outtmpl(self
):
960 outtmpl_dict
= self
.params
.get('outtmpl', {})
961 if not isinstance(outtmpl_dict
, dict):
962 outtmpl_dict
= {'default': outtmpl_dict}
963 # Remove spaces in the default template
964 if self
.params
.get('restrictfilenames'):
965 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
967 sanitize
= lambda x
: x
968 outtmpl_dict
.update({
969 k
: sanitize(v
) for k
, v
in DEFAULT_OUTTMPL
.items()
970 if outtmpl_dict
.get(k
) is None})
971 for key
, val
in outtmpl_dict
.items():
972 if isinstance(val
, bytes):
974 'Parameter outtmpl is bytes, but should be a unicode string. '
975 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
978 def get_output_path(self
, dir_type
='', filename
=None):
979 paths
= self
.params
.get('paths', {})
980 assert isinstance(paths
, dict)
982 expand_path(paths
.get('home', '').strip()),
983 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
986 # Temporary fix for #4787
987 # 'Treat' all problem characters by passing filename through preferredencoding
988 # to workaround encoding issues with subprocess on python2 @ Windows
989 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
990 path
= encodeFilename(path
, True).decode(preferredencoding())
991 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
994 def _outtmpl_expandpath(outtmpl
):
995 # expand_path translates '%%' into '%' and '$$' into '$'
996 # correspondingly that is not what we want since we need to keep
997 # '%%' intact for template dict substitution step. Working around
998 # with boundary-alike separator hack.
999 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
1000 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
1002 # outtmpl should be expand_path'ed before template dict substitution
1003 # because meta fields may contain env variables we don't want to
1004 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1005 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1006 return expand_path(outtmpl
).replace(sep
, '')
1009 def escape_outtmpl(outtmpl
):
1010 ''' Escape any remaining strings like %s, %abc% etc. '''
1012 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1013 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1017 def validate_outtmpl(cls
, outtmpl
):
1018 ''' @return None or Exception object '''
1020 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljqBUDS]'),
1021 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1022 cls
._outtmpl
_expandpath
(outtmpl
))
1024 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1026 except ValueError as err
:
1030 def _copy_infodict(info_dict
):
1031 info_dict
= dict(info_dict
)
1032 for key
in ('__original_infodict', '__postprocessors'):
1033 info_dict
.pop(key
, None)
1036 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1037 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1038 @param sanitize Whether to sanitize the output as a filename.
1039 For backward compatibility, a function can also be passed
1042 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1044 info_dict
= self
._copy
_infodict
(info_dict
)
1045 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1046 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1047 if info_dict
.get('duration', None) is not None
1049 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
1050 info_dict
['video_autonumber'] = self
._num
_videos
1051 if info_dict
.get('resolution') is None:
1052 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1054 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1055 # of %(field)s to %(field)0Nd for backward compatibility
1056 field_size_compat_map
= {
1057 'playlist_index': number_of_digits(info_dict
.get('_last_playlist_index') or 0),
1058 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1059 'autonumber': self
.params
.get('autonumber_size') or 5,
1063 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljqBUDS]'))
1068 # Field is of the form key1.key2...
1069 # where keys (except first) can be string, int or slice
1070 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1071 MATH_FIELD_RE
= r
'''(?:{field}|{num})'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
1072 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1073 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
1076 (?P<maths>(?:{math_op}{math_field})*)
1077 (?:>(?P<strf_format>.+?))?
1078 (?P<alternate>(?<!\\),[^|&)]+)?
1079 (?:&(?P<replacement>.*?))?
1080 (?:\|(?P<default>.*?))?
1081 $'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
1083 def _traverse_infodict(k
):
1087 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1089 def get_value(mdict
):
1091 value
= _traverse_infodict(mdict
['fields'])
1094 value
= float_or_none(value
)
1095 if value
is not None:
1098 offset_key
= mdict
['maths']
1100 value
= float_or_none(value
)
1104 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1105 offset_key
).group(0)
1106 offset_key
= offset_key
[len(item
):]
1107 if operator
is None:
1108 operator
= MATH_FUNCTIONS
[item
]
1110 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1111 offset
= float_or_none(item
)
1113 offset
= float_or_none(_traverse_infodict(item
))
1115 value
= operator(value
, multiplier
* offset
)
1116 except (TypeError, ZeroDivisionError):
1119 # Datetime formatting
1120 if mdict
['strf_format']:
1121 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1125 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1127 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1128 return sanitize_filename(str(value
), restricted
=restricted
,
1129 is_id
=re
.search(r
'(^|[_.])id(\.|$)', key
))
1131 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1132 sanitize
= bool(sanitize
)
1134 def _dumpjson_default(obj
):
1135 if isinstance(obj
, (set, LazyList
)):
1139 def create_key(outer_mobj
):
1140 if not outer_mobj
.group('has_key'):
1141 return outer_mobj
.group(0)
1142 key
= outer_mobj
.group('key')
1143 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1144 initial_field
= mobj
.group('fields') if mobj
else ''
1145 value
, replacement
, default
= None, None, na
1147 mobj
= mobj
.groupdict()
1148 default
= mobj
['default'] if mobj
['default'] is not None else default
1149 value
= get_value(mobj
)
1150 replacement
= mobj
['replacement']
1151 if value
is None and mobj
['alternate']:
1152 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['alternate'][1:])
1156 fmt
= outer_mobj
.group('format')
1157 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1158 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
1160 value
= default
if value
is None else value
if replacement
is None else replacement
1162 flags
= outer_mobj
.group('conversion') or ''
1163 str_fmt
= f
'{fmt[:-1]}s'
1164 if fmt
[-1] == 'l': # list
1165 delim
= '\n' if '#' in flags
else ', '
1166 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1167 elif fmt
[-1] == 'j': # json
1168 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1169 elif fmt
[-1] == 'q': # quoted
1170 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1171 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1172 elif fmt
[-1] == 'B': # bytes
1173 value
= f
'%{str_fmt}'.encode('utf-8') % str(value
).encode('utf-8')
1174 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1175 elif fmt
[-1] == 'U': # unicode normalized
1176 value
, fmt
= unicodedata
.normalize(
1177 # "+" = compatibility equivalence, "#" = NFD
1178 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1180 elif fmt
[-1] == 'D': # decimal suffix
1181 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1182 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1183 factor
=1024 if '#' in flags
else 1000)
1184 elif fmt
[-1] == 'S': # filename sanitization
1185 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1186 elif fmt
[-1] == 'c':
1188 value
= str(value
)[0]
1191 elif fmt
[-1] not in 'rs': # numeric
1192 value
= float_or_none(value
)
1194 value
, fmt
= default
, 's'
1198 # If value is an object, sanitize might convert it to a string
1199 # So we convert it to repr first
1200 value
, fmt
= repr(value
), str_fmt
1201 if fmt
[-1] in 'csr':
1202 value
= sanitizer(initial_field
, value
)
1204 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1205 TMPL_DICT
[key
] = value
1206 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1208 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1210 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1211 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1212 return self
.escape_outtmpl(outtmpl
) % info_dict
1214 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
1216 outtmpl
= self
._outtmpl
_expandpath
(self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default']))
1217 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1219 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
1220 if filename
and force_ext
is not None:
1221 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1223 # https://github.com/blackjack4494/youtube-dlc/issues/85
1224 trim_file_name
= self
.params
.get('trim_file_name', False)
1226 no_ext
, *ext
= filename
.rsplit('.', 2)
1227 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1230 except ValueError as err
:
1231 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1234 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
1235 """Generate the output filename."""
1237 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
1238 if not filename
and dir_type
not in ('', 'temp'):
1242 if not self
.params
.get('paths'):
1244 elif filename
== '-':
1245 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1246 elif os
.path
.isabs(filename
):
1247 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1248 if filename
== '-' or not filename
:
1251 return self
.get_output_path(dir_type
, filename
)
1253 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1254 """ Returns None if the file should be downloaded """
1256 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1259 if 'title' in info_dict
:
1260 # This can happen when we're just evaluating the playlist
1261 title
= info_dict
['title']
1262 matchtitle
= self
.params
.get('matchtitle', False)
1264 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1265 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1266 rejecttitle
= self
.params
.get('rejecttitle', False)
1268 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1269 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1270 date
= info_dict
.get('upload_date')
1271 if date
is not None:
1272 dateRange
= self
.params
.get('daterange', DateRange())
1273 if date
not in dateRange
:
1274 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1275 view_count
= info_dict
.get('view_count')
1276 if view_count
is not None:
1277 min_views
= self
.params
.get('min_views')
1278 if min_views
is not None and view_count
< min_views
:
1279 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1280 max_views
= self
.params
.get('max_views')
1281 if max_views
is not None and view_count
> max_views
:
1282 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1283 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1284 return 'Skipping "%s" because it is age restricted' % video_title
1286 match_filter
= self
.params
.get('match_filter')
1287 if match_filter
is not None:
1289 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1291 # For backward compatibility
1292 ret
= None if incomplete
else match_filter(info_dict
)
1297 if self
.in_download_archive(info_dict
):
1298 reason
= '%s has already been recorded in the archive' % video_title
1299 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1301 reason
= check_filter()
1302 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1303 if reason
is not None:
1305 self
.to_screen('[download] ' + reason
)
1306 if self
.params
.get(break_opt
, False):
1311 def add_extra_info(info_dict
, extra_info
):
1312 '''Set the keys from extra_info in info dict if they are missing'''
1313 for key
, value
in extra_info
.items():
1314 info_dict
.setdefault(key
, value
)
1316 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1317 process
=True, force_generic_extractor
=False):
1319 Return a list with a dictionary for each video extracted.
1322 url -- URL to extract
1325 download -- whether to download videos during extraction
1326 ie_key -- extractor key hint
1327 extra_info -- dictionary containing the extra values to add to each result
1328 process -- whether to resolve all unresolved references (URLs, playlist items),
1329 must be True for download to work.
1330 force_generic_extractor -- force using the generic extractor
1333 if extra_info
is None:
1336 if not ie_key
and force_generic_extractor
:
1340 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1344 for ie_key
, ie
in ies
.items():
1345 if not ie
.suitable(url
):
1348 if not ie
.working():
1349 self
.report_warning('The program functionality for this site has been marked as broken, '
1350 'and will probably not work.')
1352 temp_id
= ie
.get_temp_id(url
)
1353 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1354 self
.to_screen(f
'[{ie_key}] {temp_id}: has already been recorded in the archive')
1355 if self
.params
.get('break_on_existing', False):
1356 raise ExistingVideoReached()
1358 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1360 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1362 def __handle_extraction_exceptions(func
):
1363 @functools.wraps(func
)
1364 def wrapper(self
, *args
, **kwargs
):
1367 return func(self
, *args
, **kwargs
)
1368 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1370 except ReExtractInfo
as e
:
1372 self
.to_screen(f
'{e}; Re-extracting data')
1374 self
.to_stderr('\r')
1375 self
.report_warning(f
'{e}; Re-extracting data')
1377 except GeoRestrictedError
as e
:
1380 msg
+= '\nThis video is available in %s.' % ', '.join(
1381 map(ISO3166Utils
.short2full
, e
.countries
))
1382 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1383 self
.report_error(msg
)
1384 except ExtractorError
as e
: # An error we somewhat expected
1385 self
.report_error(str(e
), e
.format_traceback())
1386 except Exception as e
:
1387 if self
.params
.get('ignoreerrors'):
1388 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1394 def _wait_for_video(self
, ie_result
):
1395 if (not self
.params
.get('wait_for_video')
1396 or ie_result
.get('_type', 'video') != 'video'
1397 or ie_result
.get('formats') or ie_result
.get('url')):
1400 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1405 self
.to_screen(msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r', skip_eol
=True)
1408 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1409 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1410 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1411 diff
= random
.randrange(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
)
1412 self
.report_warning('Release time of video is not known')
1413 elif (diff
or 0) <= 0:
1414 self
.report_warning('Video should already be available according to extracted info')
1415 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1416 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1418 wait_till
= time
.time() + diff
1421 diff
= wait_till
- time
.time()
1424 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1425 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1427 except KeyboardInterrupt:
1429 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1430 except BaseException
as e
:
1431 if not isinstance(e
, ReExtractInfo
):
1435 @__handle_extraction_exceptions
1436 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1437 ie_result
= ie
.extract(url
)
1438 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1440 if isinstance(ie_result
, list):
1441 # Backwards compatibility: old IE result format
1443 '_type': 'compat_list',
1444 'entries': ie_result
,
1446 if extra_info
.get('original_url'):
1447 ie_result
.setdefault('original_url', extra_info
['original_url'])
1448 self
.add_default_extra_info(ie_result
, ie
, url
)
1450 self
._wait
_for
_video
(ie_result
)
1451 return self
.process_ie_result(ie_result
, download
, extra_info
)
1455 def add_default_extra_info(self
, ie_result
, ie
, url
):
1457 self
.add_extra_info(ie_result
, {
1459 'original_url': url
,
1460 'webpage_url_basename': url_basename(url
),
1461 'webpage_url_domain': get_domain(url
),
1464 self
.add_extra_info(ie_result
, {
1465 'extractor': ie
.IE_NAME
,
1466 'extractor_key': ie
.ie_key(),
1469 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1471 Take the result of the ie(may be modified) and resolve all unresolved
1472 references (URLs, playlist items).
1474 It will also download the videos if 'download'.
1475 Returns the resolved ie_result.
1477 if extra_info
is None:
1479 result_type
= ie_result
.get('_type', 'video')
1481 if result_type
in ('url', 'url_transparent'):
1482 ie_result
['url'] = sanitize_url(ie_result
['url'])
1483 if ie_result
.get('original_url'):
1484 extra_info
.setdefault('original_url', ie_result
['original_url'])
1486 extract_flat
= self
.params
.get('extract_flat', False)
1487 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1488 or extract_flat
is True):
1489 info_copy
= ie_result
.copy()
1490 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1491 if ie
and not ie_result
.get('id'):
1492 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1493 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1494 self
.add_extra_info(info_copy
, extra_info
)
1495 info_copy
, _
= self
.pre_process(info_copy
)
1496 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1497 if self
.params
.get('force_write_download_archive', False):
1498 self
.record_download_archive(info_copy
)
1501 if result_type
== 'video':
1502 self
.add_extra_info(ie_result
, extra_info
)
1503 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1504 additional_urls
= (ie_result
or {}).get('additional_urls')
1506 # TODO: Improve MetadataParserPP to allow setting a list
1507 if isinstance(additional_urls
, compat_str
):
1508 additional_urls
= [additional_urls
]
1510 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1511 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1512 ie_result
['additional_entries'] = [
1514 url
, download
, extra_info
=extra_info
,
1515 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1516 for url
in additional_urls
1519 elif result_type
== 'url':
1520 # We have to add extra_info to the results because it may be
1521 # contained in a playlist
1522 return self
.extract_info(
1523 ie_result
['url'], download
,
1524 ie_key
=ie_result
.get('ie_key'),
1525 extra_info
=extra_info
)
1526 elif result_type
== 'url_transparent':
1527 # Use the information from the embedding page
1528 info
= self
.extract_info(
1529 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1530 extra_info
=extra_info
, download
=False, process
=False)
1532 # extract_info may return None when ignoreerrors is enabled and
1533 # extraction failed with an error, don't crash and return early
1538 force_properties
= dict(
1539 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1540 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1541 if f
in force_properties
:
1542 del force_properties
[f
]
1543 new_result
= info
.copy()
1544 new_result
.update(force_properties
)
1546 # Extracted info may not be a video result (i.e.
1547 # info.get('_type', 'video') != video) but rather an url or
1548 # url_transparent. In such cases outer metadata (from ie_result)
1549 # should be propagated to inner one (info). For this to happen
1550 # _type of info should be overridden with url_transparent. This
1551 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1552 if new_result
.get('_type') == 'url':
1553 new_result
['_type'] = 'url_transparent'
1555 return self
.process_ie_result(
1556 new_result
, download
=download
, extra_info
=extra_info
)
1557 elif result_type
in ('playlist', 'multi_video'):
1558 # Protect from infinite recursion due to recursively nested playlists
1559 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1560 webpage_url
= ie_result
['webpage_url']
1561 if webpage_url
in self
._playlist
_urls
:
1563 '[download] Skipping already downloaded playlist: %s'
1564 % ie_result
.get('title') or ie_result
.get('id'))
1567 self
._playlist
_level
+= 1
1568 self
._playlist
_urls
.add(webpage_url
)
1569 self
._sanitize
_thumbnails
(ie_result
)
1571 return self
.__process
_playlist
(ie_result
, download
)
1573 self
._playlist
_level
-= 1
1574 if not self
._playlist
_level
:
1575 self
._playlist
_urls
.clear()
1576 elif result_type
== 'compat_list':
1577 self
.report_warning(
1578 'Extractor %s returned a compat_list result. '
1579 'It needs to be updated.' % ie_result
.get('extractor'))
1582 self
.add_extra_info(r
, {
1583 'extractor': ie_result
['extractor'],
1584 'webpage_url': ie_result
['webpage_url'],
1585 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1586 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1587 'extractor_key': ie_result
['extractor_key'],
1590 ie_result
['entries'] = [
1591 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1592 for r
in ie_result
['entries']
1596 raise Exception('Invalid result type: %s' % result_type
)
1598 def _ensure_dir_exists(self
, path
):
1599 return make_dir(path
, self
.report_error
)
1602 def _playlist_infodict(ie_result
, **kwargs
):
1605 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1606 'playlist_id': ie_result
.get('id'),
1607 'playlist_title': ie_result
.get('title'),
1608 'playlist_uploader': ie_result
.get('uploader'),
1609 'playlist_uploader_id': ie_result
.get('uploader_id'),
1610 'playlist_index': 0,
1614 def __process_playlist(self
, ie_result
, download
):
1615 # We process each entry in the playlist
1616 playlist
= ie_result
.get('title') or ie_result
.get('id')
1617 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1619 if 'entries' not in ie_result
:
1620 raise EntryNotInPlaylist('There are no entries')
1622 MissingEntry
= object()
1623 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1624 if incomplete_entries
:
1625 def fill_missing_entries(entries
, indices
):
1626 ret
= [MissingEntry
] * max(indices
)
1627 for i
, entry
in zip(indices
, entries
):
1630 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1632 playlist_results
= []
1634 playliststart
= self
.params
.get('playliststart', 1)
1635 playlistend
= self
.params
.get('playlistend')
1636 # For backwards compatibility, interpret -1 as whole list
1637 if playlistend
== -1:
1640 playlistitems_str
= self
.params
.get('playlist_items')
1641 playlistitems
= None
1642 if playlistitems_str
is not None:
1643 def iter_playlistitems(format
):
1644 for string_segment
in format
.split(','):
1645 if '-' in string_segment
:
1646 start
, end
= string_segment
.split('-')
1647 for item
in range(int(start
), int(end
) + 1):
1650 yield int(string_segment
)
1651 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1653 ie_entries
= ie_result
['entries']
1654 if isinstance(ie_entries
, list):
1655 playlist_count
= len(ie_entries
)
1656 msg
= f
'Collected {playlist_count} videos; downloading %d of them'
1657 ie_result
['playlist_count'] = ie_result
.get('playlist_count') or playlist_count
1660 return ie_entries
[i
- 1]
1662 msg
= 'Downloading %d videos'
1663 if not isinstance(ie_entries
, (PagedList
, LazyList
)):
1664 ie_entries
= LazyList(ie_entries
)
1667 return YoutubeDL
.__handle
_extraction
_exceptions
(
1668 lambda self
, i
: ie_entries
[i
- 1]
1671 entries
, broken
= [], False
1672 items
= playlistitems
if playlistitems
is not None else itertools
.count(playliststart
)
1676 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1680 entry
= get_entry(i
)
1681 if entry
is MissingEntry
:
1682 raise EntryNotInPlaylist()
1683 except (IndexError, EntryNotInPlaylist
):
1684 if incomplete_entries
:
1685 raise EntryNotInPlaylist(f
'Entry {i} cannot be found')
1686 elif not playlistitems
:
1688 entries
.append(entry
)
1690 if entry
is not None:
1691 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1692 except (ExistingVideoReached
, RejectedVideoReached
):
1695 ie_result
['entries'] = entries
1697 # Save playlist_index before re-ordering
1699 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1700 for i
, entry
in enumerate(entries
, 1)
1701 if entry
is not None]
1702 n_entries
= len(entries
)
1704 if not (ie_result
.get('playlist_count') or broken
or playlistitems
or playlistend
):
1705 ie_result
['playlist_count'] = n_entries
1707 if not playlistitems
and (playliststart
!= 1 or playlistend
):
1708 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1709 ie_result
['requested_entries'] = playlistitems
1711 _infojson_written
= False
1712 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1713 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1714 self
.list_thumbnails(ie_result
)
1715 if write_playlist_files
and not self
.params
.get('simulate'):
1716 ie_copy
= self
._playlist
_infodict
(ie_result
, n_entries
=n_entries
)
1717 _infojson_written
= self
._write
_info
_json
(
1718 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1719 if _infojson_written
is None:
1721 if self
._write
_description
('playlist', ie_result
,
1722 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1724 # TODO: This should be passed to ThumbnailsConvertor if necessary
1725 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1727 if self
.params
.get('playlistreverse', False):
1728 entries
= entries
[::-1]
1729 if self
.params
.get('playlistrandom', False):
1730 random
.shuffle(entries
)
1732 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1734 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1736 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1737 for i
, entry_tuple
in enumerate(entries
, 1):
1738 playlist_index
, entry
= entry_tuple
1739 if 'playlist-index' in self
.params
.get('compat_opts', []):
1740 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1
1741 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1742 # This __x_forwarded_for_ip thing is a bit ugly but requires
1745 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1747 'n_entries': n_entries
,
1748 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1749 'playlist_count': ie_result
.get('playlist_count'),
1750 'playlist_index': playlist_index
,
1751 'playlist_autonumber': i
,
1752 'playlist': playlist
,
1753 'playlist_id': ie_result
.get('id'),
1754 'playlist_title': ie_result
.get('title'),
1755 'playlist_uploader': ie_result
.get('uploader'),
1756 'playlist_uploader_id': ie_result
.get('uploader_id'),
1757 'extractor': ie_result
['extractor'],
1758 'webpage_url': ie_result
['webpage_url'],
1759 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1760 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1761 'extractor_key': ie_result
['extractor_key'],
1764 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1767 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1768 if not entry_result
:
1770 if failures
>= max_failures
:
1772 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1774 playlist_results
.append(entry_result
)
1775 ie_result
['entries'] = playlist_results
1777 # Write the updated info to json
1778 if _infojson_written
and self
._write
_info
_json
(
1779 'updated playlist', ie_result
,
1780 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1783 ie_result
= self
.run_all_pps('playlist', ie_result
)
1784 self
.to_screen(f
'[download] Finished downloading playlist: {playlist}')
1787 @__handle_extraction_exceptions
1788 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1789 return self
.process_ie_result(
1790 entry
, download
=download
, extra_info
=extra_info
)
1792 def _build_format_filter(self
, filter_spec
):
1793 " Returns a function to filter the formats according to the filter_spec "
1803 operator_rex
= re
.compile(r
'''(?x)\s*
1804 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1805 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1806 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1807 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1808 m
= operator_rex
.fullmatch(filter_spec
)
1811 comparison_value
= int(m
.group('value'))
1813 comparison_value
= parse_filesize(m
.group('value'))
1814 if comparison_value
is None:
1815 comparison_value
= parse_filesize(m
.group('value') + 'B')
1816 if comparison_value
is None:
1818 'Invalid value %r in format specification %r' % (
1819 m
.group('value'), filter_spec
))
1820 op
= OPERATORS
[m
.group('op')]
1825 '^=': lambda attr
, value
: attr
.startswith(value
),
1826 '$=': lambda attr
, value
: attr
.endswith(value
),
1827 '*=': lambda attr
, value
: value
in attr
,
1829 str_operator_rex
= re
.compile(r
'''(?x)\s*
1830 (?P<key>[a-zA-Z0-9._-]+)\s*
1831 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1832 (?P<value>[a-zA-Z0-9._-]+)\s*
1833 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1834 m
= str_operator_rex
.fullmatch(filter_spec
)
1836 comparison_value
= m
.group('value')
1837 str_op
= STR_OPERATORS
[m
.group('op')]
1838 if m
.group('negation'):
1839 op
= lambda attr
, value
: not str_op(attr
, value
)
1844 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1847 actual_value
= f
.get(m
.group('key'))
1848 if actual_value
is None:
1849 return m
.group('none_inclusive')
1850 return op(actual_value
, comparison_value
)
1853 def _check_formats(self
, formats
):
1855 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1856 path
= self
.get_output_path('temp')
1857 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1859 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1862 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1863 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1866 if os
.path
.exists(temp_file
.name
):
1868 os
.remove(temp_file
.name
)
1870 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1874 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1876 def _default_format_spec(self
, info_dict
, download
=True):
1879 merger
= FFmpegMergerPP(self
)
1880 return merger
.available
and merger
.can_merge()
1883 not self
.params
.get('simulate')
1887 or info_dict
.get('is_live', False)
1888 or self
.outtmpl_dict
['default'] == '-'))
1891 or self
.params
.get('allow_multiple_audio_streams', False)
1892 or 'format-spec' in self
.params
.get('compat_opts', []))
1895 'best/bestvideo+bestaudio' if prefer_best
1896 else 'bestvideo*+bestaudio/best' if not compat
1897 else 'bestvideo+bestaudio/best')
1899 def build_format_selector(self
, format_spec
):
1900 def syntax_error(note
, start
):
1902 'Invalid format specification: '
1903 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1904 return SyntaxError(message
)
1906 PICKFIRST
= 'PICKFIRST'
1910 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1912 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1913 'video': self
.params
.get('allow_multiple_video_streams', False)}
1915 check_formats
= self
.params
.get('check_formats') == 'selected'
1917 def _parse_filter(tokens
):
1919 for type, string
, start
, _
, _
in tokens
:
1920 if type == tokenize
.OP
and string
== ']':
1921 return ''.join(filter_parts
)
1923 filter_parts
.append(string
)
1925 def _remove_unused_ops(tokens
):
1926 # Remove operators that we don't use and join them with the surrounding strings
1927 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1928 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1929 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1930 for type, string
, start
, end
, line
in tokens
:
1931 if type == tokenize
.OP
and string
== '[':
1933 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1935 yield type, string
, start
, end
, line
1936 # everything inside brackets will be handled by _parse_filter
1937 for type, string
, start
, end
, line
in tokens
:
1938 yield type, string
, start
, end
, line
1939 if type == tokenize
.OP
and string
== ']':
1941 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1943 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1945 yield type, string
, start
, end
, line
1946 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1948 last_string
= string
1952 last_string
+= string
1954 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1956 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1958 current_selector
= None
1959 for type, string
, start
, _
, _
in tokens
:
1960 # ENCODING is only defined in python 3.x
1961 if type == getattr(tokenize
, 'ENCODING', None):
1963 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1964 current_selector
= FormatSelector(SINGLE
, string
, [])
1965 elif type == tokenize
.OP
:
1967 if not inside_group
:
1968 # ')' will be handled by the parentheses group
1969 tokens
.restore_last_token()
1971 elif inside_merge
and string
in ['/', ',']:
1972 tokens
.restore_last_token()
1974 elif inside_choice
and string
== ',':
1975 tokens
.restore_last_token()
1978 if not current_selector
:
1979 raise syntax_error('"," must follow a format selector', start
)
1980 selectors
.append(current_selector
)
1981 current_selector
= None
1983 if not current_selector
:
1984 raise syntax_error('"/" must follow a format selector', start
)
1985 first_choice
= current_selector
1986 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1987 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1989 if not current_selector
:
1990 current_selector
= FormatSelector(SINGLE
, 'best', [])
1991 format_filter
= _parse_filter(tokens
)
1992 current_selector
.filters
.append(format_filter
)
1994 if current_selector
:
1995 raise syntax_error('Unexpected "("', start
)
1996 group
= _parse_format_selection(tokens
, inside_group
=True)
1997 current_selector
= FormatSelector(GROUP
, group
, [])
1999 if not current_selector
:
2000 raise syntax_error('Unexpected "+"', start
)
2001 selector_1
= current_selector
2002 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2004 raise syntax_error('Expected a selector', start
)
2005 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2007 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
2008 elif type == tokenize
.ENDMARKER
:
2010 if current_selector
:
2011 selectors
.append(current_selector
)
2014 def _merge(formats_pair
):
2015 format_1
, format_2
= formats_pair
2018 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2019 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2021 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2022 get_no_more
= {'video': False, 'audio': False}
2023 for (i
, fmt_info
) in enumerate(formats_info
):
2024 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2027 for aud_vid
in ['audio', 'video']:
2028 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2029 if get_no_more
[aud_vid
]:
2032 get_no_more
[aud_vid
] = True
2034 if len(formats_info
) == 1:
2035 return formats_info
[0]
2037 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2038 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2040 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2041 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2043 output_ext
= self
.params
.get('merge_output_format')
2046 output_ext
= the_only_video
['ext']
2047 elif the_only_audio
and not video_fmts
:
2048 output_ext
= the_only_audio
['ext']
2052 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2055 'requested_formats': formats_info
,
2056 'format': '+'.join(filtered('format')),
2057 'format_id': '+'.join(filtered('format_id')),
2059 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2060 'language': '+'.join(orderedSet(filtered('language'))) or None,
2061 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2062 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2063 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2068 'width': the_only_video
.get('width'),
2069 'height': the_only_video
.get('height'),
2070 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2071 'fps': the_only_video
.get('fps'),
2072 'dynamic_range': the_only_video
.get('dynamic_range'),
2073 'vcodec': the_only_video
.get('vcodec'),
2074 'vbr': the_only_video
.get('vbr'),
2075 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2080 'acodec': the_only_audio
.get('acodec'),
2081 'abr': the_only_audio
.get('abr'),
2082 'asr': the_only_audio
.get('asr'),
2087 def _check_formats(formats
):
2088 if not check_formats
:
2091 yield from self
._check
_formats
(formats
)
2093 def _build_selector_function(selector
):
2094 if isinstance(selector
, list): # ,
2095 fs
= [_build_selector_function(s
) for s
in selector
]
2097 def selector_function(ctx
):
2100 return selector_function
2102 elif selector
.type == GROUP
: # ()
2103 selector_function
= _build_selector_function(selector
.selector
)
2105 elif selector
.type == PICKFIRST
: # /
2106 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2108 def selector_function(ctx
):
2110 picked_formats
= list(f(ctx
))
2112 return picked_formats
2115 elif selector
.type == MERGE
: # +
2116 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2118 def selector_function(ctx
):
2119 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2122 elif selector
.type == SINGLE
: # atom
2123 format_spec
= selector
.selector
or 'best'
2125 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2126 if format_spec
== 'all':
2127 def selector_function(ctx
):
2128 yield from _check_formats(ctx
['formats'][::-1])
2129 elif format_spec
== 'mergeall':
2130 def selector_function(ctx
):
2131 formats
= list(_check_formats(ctx
['formats']))
2134 merged_format
= formats
[-1]
2135 for f
in formats
[-2::-1]:
2136 merged_format
= _merge((merged_format
, f
))
2140 format_fallback
, format_reverse
, format_idx
= False, True, 1
2142 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2144 if mobj
is not None:
2145 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2146 format_reverse
= mobj
.group('bw')[0] == 'b'
2147 format_type
= (mobj
.group('type') or [None])[0]
2148 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2149 format_modified
= mobj
.group('mod') is not None
2151 format_fallback
= not format_type
and not format_modified
# for b, w
2153 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2154 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2155 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2156 if format_type
# bv, ba, wv, wa
2157 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2158 if not format_modified
# b, w
2159 else lambda f
: True) # b*, w*
2160 filter_f
= lambda f
: _filter_f(f
) and (
2161 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2163 if format_spec
in self
._format
_selection
_exts
['audio']:
2164 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2165 elif format_spec
in self
._format
_selection
_exts
['video']:
2166 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2167 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2168 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2170 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2172 def selector_function(ctx
):
2173 formats
= list(ctx
['formats'])
2174 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2175 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
2176 # for extractors with incomplete formats (audio only (soundcloud)
2177 # or video only (imgur)) best/worst will fallback to
2178 # best/worst {video,audio}-only format
2180 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2182 yield matches
[format_idx
- 1]
2186 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2188 def final_selector(ctx
):
2189 ctx_copy
= dict(ctx
)
2190 for _filter
in filters
:
2191 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2192 return selector_function(ctx_copy
)
2193 return final_selector
2195 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
2197 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
2198 except tokenize
.TokenError
:
2199 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2201 class TokenIterator(object):
2202 def __init__(self
, tokens
):
2203 self
.tokens
= tokens
2210 if self
.counter
>= len(self
.tokens
):
2211 raise StopIteration()
2212 value
= self
.tokens
[self
.counter
]
2218 def restore_last_token(self
):
2221 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2222 return _build_selector_function(parsed_selector
)
2224 def _calc_headers(self
, info_dict
):
2225 res
= std_headers
.copy()
2227 add_headers
= info_dict
.get('http_headers')
2229 res
.update(add_headers
)
2231 cookies
= self
._calc
_cookies
(info_dict
)
2233 res
['Cookie'] = cookies
2235 if 'X-Forwarded-For' not in res
:
2236 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2237 if x_forwarded_for_ip
:
2238 res
['X-Forwarded-For'] = x_forwarded_for_ip
2242 def _calc_cookies(self
, info_dict
):
2243 pr
= sanitized_Request(info_dict
['url'])
2244 self
.cookiejar
.add_cookie_header(pr
)
2245 return pr
.get_header('Cookie')
2247 def _sort_thumbnails(self
, thumbnails
):
2248 thumbnails
.sort(key
=lambda t
: (
2249 t
.get('preference') if t
.get('preference') is not None else -1,
2250 t
.get('width') if t
.get('width') is not None else -1,
2251 t
.get('height') if t
.get('height') is not None else -1,
2252 t
.get('id') if t
.get('id') is not None else '',
2255 def _sanitize_thumbnails(self
, info_dict
):
2256 thumbnails
= info_dict
.get('thumbnails')
2257 if thumbnails
is None:
2258 thumbnail
= info_dict
.get('thumbnail')
2260 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2264 def check_thumbnails(thumbnails
):
2265 for t
in thumbnails
:
2266 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2268 self
.urlopen(HEADRequest(t
['url']))
2269 except network_exceptions
as err
:
2270 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2274 self
._sort
_thumbnails
(thumbnails
)
2275 for i
, t
in enumerate(thumbnails
):
2276 if t
.get('id') is None:
2278 if t
.get('width') and t
.get('height'):
2279 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2280 t
['url'] = sanitize_url(t
['url'])
2282 if self
.params
.get('check_formats') is True:
2283 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2285 info_dict
['thumbnails'] = thumbnails
2287 def process_video_result(self
, info_dict
, download
=True):
2288 assert info_dict
.get('_type', 'video') == 'video'
2289 self
._num
_videos
+= 1
2291 if 'id' not in info_dict
:
2292 raise ExtractorError('Missing "id" field in extractor result')
2293 if 'title' not in info_dict
:
2294 raise ExtractorError('Missing "title" field in extractor result',
2295 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2297 def report_force_conversion(field
, field_not
, conversion
):
2298 self
.report_warning(
2299 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2300 % (field
, field_not
, conversion
))
2302 def sanitize_string_field(info
, string_field
):
2303 field
= info
.get(string_field
)
2304 if field
is None or isinstance(field
, compat_str
):
2306 report_force_conversion(string_field
, 'a string', 'string')
2307 info
[string_field
] = compat_str(field
)
2309 def sanitize_numeric_fields(info
):
2310 for numeric_field
in self
._NUMERIC
_FIELDS
:
2311 field
= info
.get(numeric_field
)
2312 if field
is None or isinstance(field
, compat_numeric_types
):
2314 report_force_conversion(numeric_field
, 'numeric', 'int')
2315 info
[numeric_field
] = int_or_none(field
)
2317 sanitize_string_field(info_dict
, 'id')
2318 sanitize_numeric_fields(info_dict
)
2320 if 'playlist' not in info_dict
:
2321 # It isn't part of a playlist
2322 info_dict
['playlist'] = None
2323 info_dict
['playlist_index'] = None
2325 self
._sanitize
_thumbnails
(info_dict
)
2327 thumbnail
= info_dict
.get('thumbnail')
2328 thumbnails
= info_dict
.get('thumbnails')
2330 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2332 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2334 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2335 info_dict
['display_id'] = info_dict
['id']
2337 if info_dict
.get('duration') is not None:
2338 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2340 for ts_key
, date_key
in (
2341 ('timestamp', 'upload_date'),
2342 ('release_timestamp', 'release_date'),
2343 ('modified_timestamp', 'modified_date'),
2345 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2346 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2347 # see http://bugs.python.org/issue1646728)
2349 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2350 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2351 except (ValueError, OverflowError, OSError):
2354 live_keys
= ('is_live', 'was_live')
2355 live_status
= info_dict
.get('live_status')
2356 if live_status
is None:
2357 for key
in live_keys
:
2358 if info_dict
.get(key
) is False:
2360 if info_dict
.get(key
):
2363 if all(info_dict
.get(key
) is False for key
in live_keys
):
2364 live_status
= 'not_live'
2366 info_dict
['live_status'] = live_status
2367 for key
in live_keys
:
2368 if info_dict
.get(key
) is None:
2369 info_dict
[key
] = (live_status
== key
)
2371 # Auto generate title fields corresponding to the *_number fields when missing
2372 # in order to always have clean titles. This is very common for TV series.
2373 for field
in ('chapter', 'season', 'episode'):
2374 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2375 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2377 for cc_kind
in ('subtitles', 'automatic_captions'):
2378 cc
= info_dict
.get(cc_kind
)
2380 for _
, subtitle
in cc
.items():
2381 for subtitle_format
in subtitle
:
2382 if subtitle_format
.get('url'):
2383 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2384 if subtitle_format
.get('ext') is None:
2385 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2387 automatic_captions
= info_dict
.get('automatic_captions')
2388 subtitles
= info_dict
.get('subtitles')
2390 info_dict
['requested_subtitles'] = self
.process_subtitles(
2391 info_dict
['id'], subtitles
, automatic_captions
)
2393 if info_dict
.get('formats') is None:
2394 # There's only one format available
2395 formats
= [info_dict
]
2397 formats
= info_dict
['formats']
2399 info_dict
['__has_drm'] = any(f
.get('has_drm') for f
in formats
)
2400 if not self
.params
.get('allow_unplayable_formats'):
2401 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2403 # backward compatibility
2404 info_dict
['fulltitle'] = info_dict
['title']
2406 if info_dict
.get('is_live'):
2407 get_from_start
= bool(self
.params
.get('live_from_start'))
2408 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2409 if not get_from_start
:
2410 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2413 self
.raise_no_formats(info_dict
)
2415 def is_wellformed(f
):
2418 self
.report_warning(
2419 '"url" field is missing or empty - skipping format, '
2420 'there is an error in extractor')
2422 if isinstance(url
, bytes):
2423 sanitize_string_field(f
, 'url')
2426 # Filter out malformed formats for better extraction robustness
2427 formats
= list(filter(is_wellformed
, formats
))
2431 # We check that all the formats have the format and format_id fields
2432 for i
, format
in enumerate(formats
):
2433 sanitize_string_field(format
, 'format_id')
2434 sanitize_numeric_fields(format
)
2435 format
['url'] = sanitize_url(format
['url'])
2436 if not format
.get('format_id'):
2437 format
['format_id'] = compat_str(i
)
2439 # Sanitize format_id from characters used in format selector expression
2440 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2441 format_id
= format
['format_id']
2442 if format_id
not in formats_dict
:
2443 formats_dict
[format_id
] = []
2444 formats_dict
[format_id
].append(format
)
2446 # Make sure all formats have unique format_id
2447 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2448 for format_id
, ambiguous_formats
in formats_dict
.items():
2449 ambigious_id
= len(ambiguous_formats
) > 1
2450 for i
, format
in enumerate(ambiguous_formats
):
2452 format
['format_id'] = '%s-%d' % (format_id
, i
)
2453 if format
.get('ext') is None:
2454 format
['ext'] = determine_ext(format
['url']).lower()
2455 # Ensure there is no conflict between id and ext in format selection
2456 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2457 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2458 format
['format_id'] = 'f%s' % format
['format_id']
2460 for i
, format
in enumerate(formats
):
2461 if format
.get('format') is None:
2462 format
['format'] = '{id} - {res}{note}'.format(
2463 id=format
['format_id'],
2464 res
=self
.format_resolution(format
),
2465 note
=format_field(format
, 'format_note', ' (%s)'),
2467 if format
.get('protocol') is None:
2468 format
['protocol'] = determine_protocol(format
)
2469 if format
.get('resolution') is None:
2470 format
['resolution'] = self
.format_resolution(format
, default
=None)
2471 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2472 format
['dynamic_range'] = 'SDR'
2473 if (info_dict
.get('duration') and format
.get('tbr')
2474 and not format
.get('filesize') and not format
.get('filesize_approx')):
2475 format
['filesize_approx'] = info_dict
['duration'] * format
['tbr'] * (1024 / 8)
2477 # Add HTTP headers, so that external programs can use them from the
2479 full_format_info
= info_dict
.copy()
2480 full_format_info
.update(format
)
2481 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2482 # Remove private housekeeping stuff
2483 if '__x_forwarded_for_ip' in info_dict
:
2484 del info_dict
['__x_forwarded_for_ip']
2486 # TODO Central sorting goes here
2488 if self
.params
.get('check_formats') is True:
2489 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2491 if not formats
or formats
[0] is not info_dict
:
2492 # only set the 'formats' fields if the original info_dict list them
2493 # otherwise we end up with a circular reference, the first (and unique)
2494 # element in the 'formats' field in info_dict is info_dict itself,
2495 # which can't be exported to json
2496 info_dict
['formats'] = formats
2498 info_dict
, _
= self
.pre_process(info_dict
)
2500 # The pre-processors may have modified the formats
2501 formats
= info_dict
.get('formats', [info_dict
])
2503 list_only
= self
.params
.get('simulate') is None and (
2504 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2505 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2506 if self
.params
.get('list_thumbnails'):
2507 self
.list_thumbnails(info_dict
)
2508 if self
.params
.get('listsubtitles'):
2509 if 'automatic_captions' in info_dict
:
2510 self
.list_subtitles(
2511 info_dict
['id'], automatic_captions
, 'automatic captions')
2512 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2513 if self
.params
.get('listformats') or interactive_format_selection
:
2514 self
.list_formats(info_dict
)
2516 # Without this printing, -F --print-json will not work
2517 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2520 format_selector
= self
.format_selector
2521 if format_selector
is None:
2522 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2523 self
.write_debug('Default format spec: %s' % req_format
)
2524 format_selector
= self
.build_format_selector(req_format
)
2527 if interactive_format_selection
:
2529 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2531 format_selector
= self
.build_format_selector(req_format
)
2532 except SyntaxError as err
:
2533 self
.report_error(err
, tb
=False, is_error
=False)
2536 # While in format selection we may need to have an access to the original
2537 # format set in order to calculate some metrics or do some processing.
2538 # For now we need to be able to guess whether original formats provided
2539 # by extractor are incomplete or not (i.e. whether extractor provides only
2540 # video-only or audio-only formats) for proper formats selection for
2541 # extractors with such incomplete formats (see
2542 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2543 # Since formats may be filtered during format selection and may not match
2544 # the original formats the results may be incorrect. Thus original formats
2545 # or pre-calculated metrics should be passed to format selection routines
2547 # We will pass a context object containing all necessary additional data
2548 # instead of just formats.
2549 # This fixes incorrect format selection issue (see
2550 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2551 incomplete_formats
= (
2552 # All formats are video-only or
2553 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2554 # all formats are audio-only
2555 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2559 'incomplete_formats': incomplete_formats
,
2562 formats_to_download
= list(format_selector(ctx
))
2563 if interactive_format_selection
and not formats_to_download
:
2564 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2568 if not formats_to_download
:
2569 if not self
.params
.get('ignore_no_formats_error'):
2570 raise ExtractorError('Requested format is not available', expected
=True,
2571 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2572 self
.report_warning('Requested format is not available')
2573 # Process what we can, even without any available formats.
2574 formats_to_download
= [{}]
2576 best_format
= formats_to_download
[-1]
2580 f
'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2581 + ', '.join([f
['format_id'] for f
in formats_to_download
]))
2582 max_downloads_reached
= False
2583 for i
, fmt
in enumerate(formats_to_download
):
2584 formats_to_download
[i
] = new_info
= dict(info_dict
)
2585 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2586 new_info
.update(fmt
)
2587 new_info
['__original_infodict'] = info_dict
2589 self
.process_info(new_info
)
2590 except MaxDownloadsReached
:
2591 max_downloads_reached
= True
2592 new_info
.pop('__original_infodict')
2593 # Remove copied info
2594 for key
, val
in tuple(new_info
.items()):
2595 if info_dict
.get(key
) == val
:
2597 if max_downloads_reached
:
2600 write_archive
= set(f
.get('__write_download_archive', False) for f
in formats_to_download
)
2601 assert write_archive
.issubset({True, False, 'ignore'}
)
2602 if True in write_archive
and False not in write_archive
:
2603 self
.record_download_archive(info_dict
)
2605 info_dict
['requested_downloads'] = formats_to_download
2606 info_dict
= self
.run_all_pps('after_video', info_dict
)
2607 if max_downloads_reached
:
2608 raise MaxDownloadsReached()
2610 # We update the info dict with the selected best quality format (backwards compatibility)
2611 info_dict
.update(best_format
)
2614 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2615 """Select the requested subtitles and their format"""
2617 if normal_subtitles
and self
.params
.get('writesubtitles'):
2618 available_subs
.update(normal_subtitles
)
2619 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2620 for lang
, cap_info
in automatic_captions
.items():
2621 if lang
not in available_subs
:
2622 available_subs
[lang
] = cap_info
2624 if (not self
.params
.get('writesubtitles') and not
2625 self
.params
.get('writeautomaticsub') or not
2629 all_sub_langs
= available_subs
.keys()
2630 if self
.params
.get('allsubtitles', False):
2631 requested_langs
= all_sub_langs
2632 elif self
.params
.get('subtitleslangs', False):
2633 # A list is used so that the order of languages will be the same as
2634 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2635 requested_langs
= []
2636 for lang_re
in self
.params
.get('subtitleslangs'):
2637 if lang_re
== 'all':
2638 requested_langs
.extend(all_sub_langs
)
2640 discard
= lang_re
[0] == '-'
2642 lang_re
= lang_re
[1:]
2643 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2645 for lang
in current_langs
:
2646 while lang
in requested_langs
:
2647 requested_langs
.remove(lang
)
2649 requested_langs
.extend(current_langs
)
2650 requested_langs
= orderedSet(requested_langs
)
2651 elif 'en' in available_subs
:
2652 requested_langs
= ['en']
2654 requested_langs
= [list(all_sub_langs
)[0]]
2656 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2658 formats_query
= self
.params
.get('subtitlesformat', 'best')
2659 formats_preference
= formats_query
.split('/') if formats_query
else []
2661 for lang
in requested_langs
:
2662 formats
= available_subs
.get(lang
)
2664 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2666 for ext
in formats_preference
:
2670 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2676 self
.report_warning(
2677 'No subtitle format found matching "%s" for language %s, '
2678 'using %s' % (formats_query
, lang
, f
['ext']))
2682 def _forceprint(self
, tmpl
, info_dict
):
2683 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2684 if mobj
and mobj
.group(1):
2685 tmpl
= f
'{tmpl[:-1]} = %({tmpl[:-1]})s'
2687 tmpl
= '%({})s'.format(tmpl
)
2689 info_dict
= info_dict
.copy()
2690 info_dict
['formats_table'] = self
.render_formats_table(info_dict
)
2691 info_dict
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2692 info_dict
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2693 info_dict
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2694 self
.to_stdout(self
.evaluate_outtmpl(tmpl
, info_dict
))
2696 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2697 def print_mandatory(field
, actual_field
=None):
2698 if actual_field
is None:
2699 actual_field
= field
2700 if (self
.params
.get('force%s' % field
, False)
2701 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2702 self
.to_stdout(info_dict
[actual_field
])
2704 def print_optional(field
):
2705 if (self
.params
.get('force%s' % field
, False)
2706 and info_dict
.get(field
) is not None):
2707 self
.to_stdout(info_dict
[field
])
2709 info_dict
= info_dict
.copy()
2710 if filename
is not None:
2711 info_dict
['filename'] = filename
2712 if info_dict
.get('requested_formats') is not None:
2713 # For RTMP URLs, also include the playpath
2714 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2715 elif 'url' in info_dict
:
2716 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2718 if self
.params
['forceprint'].get('video') or self
.params
.get('forcejson'):
2719 self
.post_extract(info_dict
)
2720 for tmpl
in self
.params
['forceprint'].get('video', []):
2721 self
._forceprint
(tmpl
, info_dict
)
2723 print_mandatory('title')
2724 print_mandatory('id')
2725 print_mandatory('url', 'urls')
2726 print_optional('thumbnail')
2727 print_optional('description')
2728 print_optional('filename')
2729 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2730 self
.to_stdout(formatSeconds(info_dict
['duration']))
2731 print_mandatory('format')
2733 if self
.params
.get('forcejson'):
2734 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2736 def dl(self
, name
, info
, subtitle
=False, test
=False):
2737 if not info
.get('url'):
2738 self
.raise_no_formats(info
, True)
2741 verbose
= self
.params
.get('verbose')
2744 'quiet': self
.params
.get('quiet') or not verbose
,
2746 'noprogress': not verbose
,
2748 'skip_unavailable_fragments': False,
2749 'keep_fragments': False,
2751 '_no_ytdl_file': True,
2754 params
= self
.params
2755 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2757 for ph
in self
._progress
_hooks
:
2758 fd
.add_progress_hook(ph
)
2760 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2761 for f
in info
.get('requested_formats', []) or [info
])
2762 self
.write_debug('Invoking downloader on "%s"' % urls
)
2764 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2765 # But it may contain objects that are not deep-copyable
2766 new_info
= self
._copy
_infodict
(info
)
2767 if new_info
.get('http_headers') is None:
2768 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2769 return fd
.download(name
, new_info
, subtitle
)
2771 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2772 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2773 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2774 return existing_files
[0]
2776 for file in existing_files
:
2777 self
.report_file_delete(file)
2781 def process_info(self
, info_dict
):
2782 """Process a single resolved IE result. (Modified it in-place)"""
2784 assert info_dict
.get('_type', 'video') == 'video'
2785 original_infodict
= info_dict
2787 if 'format' not in info_dict
and 'ext' in info_dict
:
2788 info_dict
['format'] = info_dict
['ext']
2790 if self
._match
_entry
(info_dict
) is not None:
2791 info_dict
['__write_download_archive'] = 'ignore'
2794 self
.post_extract(info_dict
)
2795 self
._num
_downloads
+= 1
2797 # info_dict['_filename'] needs to be set for backward compatibility
2798 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2799 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2803 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2805 if self
.params
.get('simulate'):
2806 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2809 if full_filename
is None:
2811 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2813 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2816 if self
._write
_description
('video', info_dict
,
2817 self
.prepare_filename(info_dict
, 'description')) is None:
2820 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2821 if sub_files
is None:
2823 files_to_move
.update(dict(sub_files
))
2825 thumb_files
= self
._write
_thumbnails
(
2826 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2827 if thumb_files
is None:
2829 files_to_move
.update(dict(thumb_files
))
2831 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2832 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2833 if _infojson_written
:
2834 info_dict
['infojson_filename'] = infofn
2835 # For backward compatibility, even though it was a private field
2836 info_dict
['__infojson_filename'] = infofn
2837 elif _infojson_written
is None:
2840 # Note: Annotations are deprecated
2842 if self
.params
.get('writeannotations', False):
2843 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2845 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2847 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2848 self
.to_screen('[info] Video annotations are already present')
2849 elif not info_dict
.get('annotations'):
2850 self
.report_warning('There are no annotations to write.')
2853 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2854 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2855 annofile
.write(info_dict
['annotations'])
2856 except (KeyError, TypeError):
2857 self
.report_warning('There are no annotations to write.')
2858 except (OSError, IOError):
2859 self
.report_error('Cannot write annotations file: ' + annofn
)
2862 # Write internet shortcut files
2863 def _write_link_file(link_type
):
2864 if 'webpage_url' not in info_dict
:
2865 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2867 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2868 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
2870 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2871 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2874 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2875 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2876 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2877 template_vars
= {'url': iri_to_uri(info_dict['webpage_url'])}
2878 if link_type
== 'desktop':
2879 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2880 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
2881 except (OSError, IOError):
2882 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
2887 'url': self
.params
.get('writeurllink'),
2888 'webloc': self
.params
.get('writewebloclink'),
2889 'desktop': self
.params
.get('writedesktoplink'),
2891 if self
.params
.get('writelink'):
2892 link_type
= ('webloc' if sys
.platform
== 'darwin'
2893 else 'desktop' if sys
.platform
.startswith('linux')
2895 write_links
[link_type
] = True
2897 if any(should_write
and not _write_link_file(link_type
)
2898 for link_type
, should_write
in write_links
.items()):
2901 def replace_info_dict(new_info
):
2903 if new_info
== info_dict
:
2906 info_dict
.update(new_info
)
2909 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2910 replace_info_dict(new_info
)
2911 except PostProcessingError
as err
:
2912 self
.report_error('Preprocessing: %s' % str(err
))
2915 if self
.params
.get('skip_download'):
2916 info_dict
['filepath'] = temp_filename
2917 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2918 info_dict
['__files_to_move'] = files_to_move
2919 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
2920 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2923 info_dict
.setdefault('__postprocessors', [])
2926 def existing_video_file(*filepaths
):
2927 ext
= info_dict
.get('ext')
2928 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
2929 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
2930 default_overwrite
=False)
2932 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
2936 if info_dict
.get('requested_formats') is not None:
2938 def compatible_formats(formats
):
2939 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2940 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2941 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2942 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2946 exts
= set(format
.get('ext') for format
in formats
)
2948 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2951 for ext_sets
in COMPATIBLE_EXTS
:
2952 if ext_sets
.issuperset(exts
):
2954 # TODO: Check acodec/vcodec
2957 requested_formats
= info_dict
['requested_formats']
2958 old_ext
= info_dict
['ext']
2959 if self
.params
.get('merge_output_format') is None:
2960 if not compatible_formats(requested_formats
):
2961 info_dict
['ext'] = 'mkv'
2962 self
.report_warning(
2963 'Requested formats are incompatible for merge and will be merged into mkv')
2964 if (info_dict
['ext'] == 'webm'
2965 and info_dict
.get('thumbnails')
2966 # check with type instead of pp_key, __name__, or isinstance
2967 # since we dont want any custom PPs to trigger this
2968 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])):
2969 info_dict
['ext'] = 'mkv'
2970 self
.report_warning(
2971 'webm doesn\'t support embedding a thumbnail, mkv will be used')
2972 new_ext
= info_dict
['ext']
2974 def correct_ext(filename
, ext
=new_ext
):
2977 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2979 os
.path
.splitext(filename
)[0]
2980 if filename_real_ext
in (old_ext
, new_ext
)
2982 return '%s.%s' % (filename_wo_ext
, ext
)
2984 # Ensure filename always has a correct extension for successful merge
2985 full_filename
= correct_ext(full_filename
)
2986 temp_filename
= correct_ext(temp_filename
)
2987 dl_filename
= existing_video_file(full_filename
, temp_filename
)
2988 info_dict
['__real_download'] = False
2991 merger
= FFmpegMergerPP(self
)
2993 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
2994 if dl_filename
is not None:
2995 self
.report_file_already_downloaded(dl_filename
)
2997 for f
in requested_formats
if fd
!= FFmpegFD
else []:
2998 f
['filepath'] = fname
= prepend_extension(
2999 correct_ext(temp_filename
, info_dict
['ext']),
3000 'f%s' % f
['format_id'], info_dict
['ext'])
3001 downloaded
.append(fname
)
3002 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
3003 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3004 info_dict
['__real_download'] = real_download
3006 if self
.params
.get('allow_unplayable_formats'):
3007 self
.report_warning(
3008 'You have requested merging of multiple formats '
3009 'while also allowing unplayable formats to be downloaded. '
3010 'The formats won\'t be merged to prevent data corruption.')
3011 elif not merger
.available
:
3012 self
.report_warning(
3013 'You have requested merging of multiple formats but ffmpeg is not installed. '
3014 'The formats won\'t be merged.')
3016 if temp_filename
== '-':
3017 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3018 else 'but the formats are incompatible for simultaneous download' if merger
.available
3019 else 'but ffmpeg is not installed')
3020 self
.report_warning(
3021 f
'You have requested downloading multiple formats to stdout {reason}. '
3022 'The formats will be streamed one after the other')
3023 fname
= temp_filename
3024 for f
in requested_formats
:
3025 new_info
= dict(info_dict
)
3026 del new_info
['requested_formats']
3028 if temp_filename
!= '-':
3029 fname
= prepend_extension(
3030 correct_ext(temp_filename
, new_info
['ext']),
3031 'f%s' % f
['format_id'], new_info
['ext'])
3032 if not self
._ensure
_dir
_exists
(fname
):
3034 f
['filepath'] = fname
3035 downloaded
.append(fname
)
3036 partial_success
, real_download
= self
.dl(fname
, new_info
)
3037 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3038 success
= success
and partial_success
3040 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3041 info_dict
['__postprocessors'].append(merger
)
3042 info_dict
['__files_to_merge'] = downloaded
3043 # Even if there were no downloads, it is being merged only now
3044 info_dict
['__real_download'] = True
3046 for file in downloaded
:
3047 files_to_move
[file] = None
3049 # Just a single file
3050 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3051 if dl_filename
is None or dl_filename
== temp_filename
:
3052 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3053 # So we should try to resume the download
3054 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3055 info_dict
['__real_download'] = real_download
3057 self
.report_file_already_downloaded(dl_filename
)
3059 dl_filename
= dl_filename
or temp_filename
3060 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3062 except network_exceptions
as err
:
3063 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3065 except (OSError, IOError) as err
:
3066 raise UnavailableVideoError(err
)
3067 except (ContentTooShortError
, ) as err
:
3068 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
3071 if success
and full_filename
!= '-':
3075 fixup_policy
= self
.params
.get('fixup')
3076 vid
= info_dict
['id']
3078 if fixup_policy
in ('ignore', 'never'):
3080 elif fixup_policy
== 'warn':
3082 elif fixup_policy
!= 'force':
3083 assert fixup_policy
in ('detect_or_warn', None)
3084 if not info_dict
.get('__real_download'):
3087 def ffmpeg_fixup(cndn
, msg
, cls
):
3091 self
.report_warning(f
'{vid}: {msg}')
3095 info_dict
['__postprocessors'].append(pp
)
3097 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3099 stretched_ratio
= info_dict
.get('stretched_ratio')
3101 stretched_ratio
not in (1, None),
3102 f
'Non-uniform pixel ratio {stretched_ratio}',
3103 FFmpegFixupStretchedPP
)
3106 (info_dict
.get('requested_formats') is None
3107 and info_dict
.get('container') == 'm4a_dash'
3108 and info_dict
.get('ext') == 'm4a'),
3109 'writing DASH m4a. Only some players support this container',
3112 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3113 downloader
= downloader
.__name
__ if downloader
else None
3115 if info_dict
.get('requested_formats') is None: # Not necessary if doing merger
3116 ffmpeg_fixup(downloader
== 'HlsFD',
3117 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3119 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3120 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3122 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3123 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP
)
3127 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3128 except PostProcessingError
as err
:
3129 self
.report_error('Postprocessing: %s' % str(err
))
3132 for ph
in self
._post
_hooks
:
3133 ph(info_dict
['filepath'])
3134 except Exception as err
:
3135 self
.report_error('post hooks: %s' % str(err
))
3137 info_dict
['__write_download_archive'] = True
3139 if self
.params
.get('force_write_download_archive'):
3140 info_dict
['__write_download_archive'] = True
3142 # Make sure the info_dict was modified in-place
3143 assert info_dict
is original_infodict
3145 max_downloads
= self
.params
.get('max_downloads')
3146 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
3147 raise MaxDownloadsReached()
3149 def __download_wrapper(self
, func
):
3150 @functools.wraps(func
)
3151 def wrapper(*args
, **kwargs
):
3153 res
= func(*args
, **kwargs
)
3154 except UnavailableVideoError
as e
:
3155 self
.report_error(e
)
3156 except MaxDownloadsReached
as e
:
3157 self
.to_screen(f
'[info] {e}')
3159 except DownloadCancelled
as e
:
3160 self
.to_screen(f
'[info] {e}')
3161 if not self
.params
.get('break_per_url'):
3164 if self
.params
.get('dump_single_json', False):
3165 self
.post_extract(res
)
3166 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3169 def download(self
, url_list
):
3170 """Download a given list of URLs."""
3171 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3172 outtmpl
= self
.outtmpl_dict
['default']
3173 if (len(url_list
) > 1
3175 and '%' not in outtmpl
3176 and self
.params
.get('max_downloads') != 1):
3177 raise SameFileError(outtmpl
)
3179 for url
in url_list
:
3180 self
.__download
_wrapper
(self
.extract_info
)(
3181 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3183 return self
._download
_retcode
3185 def download_with_info_file(self
, info_filename
):
3186 with contextlib
.closing(fileinput
.FileInput(
3187 [info_filename
], mode
='r',
3188 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3189 # FileInput doesn't have a read method, we can't call json.load
3190 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3192 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3193 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3194 if not isinstance(e
, EntryNotInPlaylist
):
3195 self
.to_stderr('\r')
3196 webpage_url
= info
.get('webpage_url')
3197 if webpage_url
is not None:
3198 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3199 return self
.download([webpage_url
])
3202 return self
._download
_retcode
3205 def sanitize_info(info_dict
, remove_private_keys
=False):
3206 ''' Sanitize the infodict for converting to json '''
3207 if info_dict
is None:
3209 info_dict
.setdefault('epoch', int(time
.time()))
3210 remove_keys
= {'__original_infodict'}
# Always remove this since this may contain a copy of the entire dict
3211 keep_keys
= ['_type'] # Always keep this to facilitate load-info-json
3212 if remove_private_keys
:
3214 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3215 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3217 reject
= lambda k
, v
: k
not in keep_keys
and (
3218 k
.startswith('_') or k
in remove_keys
or v
is None)
3220 reject
= lambda k
, v
: k
in remove_keys
3223 if isinstance(obj
, dict):
3224 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3225 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3226 return list(map(filter_fn
, obj
))
3227 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3232 return filter_fn(info_dict
)
3235 def filter_requested_info(info_dict
, actually_filter
=True):
3236 ''' Alias of sanitize_info for backward compatibility '''
3237 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3240 def post_extract(info_dict
):
3241 def actual_post_extract(info_dict
):
3242 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3243 for video_dict
in info_dict
.get('entries', {}):
3244 actual_post_extract(video_dict
or {})
3247 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
3248 extra
= post_extractor().items()
3249 info_dict
.update(extra
)
3250 info_dict
.pop('__post_extractor', None)
3252 original_infodict
= info_dict
.get('__original_infodict') or {}
3253 original_infodict
.update(extra
)
3254 original_infodict
.pop('__post_extractor', None)
3256 actual_post_extract(info_dict
or {})
3258 def run_pp(self
, pp
, infodict
):
3259 files_to_delete
= []
3260 if '__files_to_move' not in infodict
:
3261 infodict
['__files_to_move'] = {}
3263 files_to_delete
, infodict
= pp
.run(infodict
)
3264 except PostProcessingError
as e
:
3265 # Must be True and not 'only_download'
3266 if self
.params
.get('ignoreerrors') is True:
3267 self
.report_error(e
)
3271 if not files_to_delete
:
3273 if self
.params
.get('keepvideo', False):
3274 for f
in files_to_delete
:
3275 infodict
['__files_to_move'].setdefault(f
, '')
3277 for old_filename
in set(files_to_delete
):
3278 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
3280 os
.remove(encodeFilename(old_filename
))
3281 except (IOError, OSError):
3282 self
.report_warning('Unable to remove downloaded original file')
3283 if old_filename
in infodict
['__files_to_move']:
3284 del infodict
['__files_to_move'][old_filename
]
3287 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3288 for tmpl
in self
.params
['forceprint'].get(key
, []):
3289 self
._forceprint
(tmpl
, info
)
3290 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3291 info
= self
.run_pp(pp
, info
)
3294 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3295 info
= dict(ie_info
)
3296 info
['__files_to_move'] = files_to_move
or {}
3297 info
= self
.run_all_pps(key
, info
)
3298 return info
, info
.pop('__files_to_move', None)
3300 def post_process(self
, filename
, info
, files_to_move
=None):
3301 """Run all the postprocessors on the given file."""
3302 info
['filepath'] = filename
3303 info
['__files_to_move'] = files_to_move
or {}
3304 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3305 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3306 del info
['__files_to_move']
3307 return self
.run_all_pps('after_move', info
)
3309 def _make_archive_id(self
, info_dict
):
3310 video_id
= info_dict
.get('id')
3313 # Future-proof against any change in case
3314 # and backwards compatibility with prior versions
3315 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3316 if extractor
is None:
3317 url
= str_or_none(info_dict
.get('url'))
3320 # Try to find matching extractor for the URL and take its ie_key
3321 for ie_key
, ie
in self
._ies
.items():
3322 if ie
.suitable(url
):
3327 return '%s %s' % (extractor
.lower(), video_id
)
3329 def in_download_archive(self
, info_dict
):
3330 fn
= self
.params
.get('download_archive')
3334 vid_id
= self
._make
_archive
_id
(info_dict
)
3336 return False # Incomplete video information
3338 return vid_id
in self
.archive
3340 def record_download_archive(self
, info_dict
):
3341 fn
= self
.params
.get('download_archive')
3344 vid_id
= self
._make
_archive
_id
(info_dict
)
3346 self
.write_debug(f
'Adding to archive: {vid_id}')
3347 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3348 archive_file
.write(vid_id
+ '\n')
3349 self
.archive
.add(vid_id
)
3352 def format_resolution(format
, default
='unknown'):
3353 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3355 if format
.get('resolution') is not None:
3356 return format
['resolution']
3357 if format
.get('width') and format
.get('height'):
3358 return '%dx%d' % (format
['width'], format
['height'])
3359 elif format
.get('height'):
3360 return '%sp' % format
['height']
3361 elif format
.get('width'):
3362 return '%dx?' % format
['width']
3365 def _list_format_headers(self
, *headers
):
3366 if self
.params
.get('listformats_table', True) is not False:
3367 return [self
._format
_screen
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3370 def _format_note(self
, fdict
):
3372 if fdict
.get('ext') in ['f4f', 'f4m']:
3373 res
+= '(unsupported)'
3374 if fdict
.get('language'):
3377 res
+= '[%s]' % fdict
['language']
3378 if fdict
.get('format_note') is not None:
3381 res
+= fdict
['format_note']
3382 if fdict
.get('tbr') is not None:
3385 res
+= '%4dk' % fdict
['tbr']
3386 if fdict
.get('container') is not None:
3389 res
+= '%s container' % fdict
['container']
3390 if (fdict
.get('vcodec') is not None
3391 and fdict
.get('vcodec') != 'none'):
3394 res
+= fdict
['vcodec']
3395 if fdict
.get('vbr') is not None:
3397 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3399 if fdict
.get('vbr') is not None:
3400 res
+= '%4dk' % fdict
['vbr']
3401 if fdict
.get('fps') is not None:
3404 res
+= '%sfps' % fdict
['fps']
3405 if fdict
.get('acodec') is not None:
3408 if fdict
['acodec'] == 'none':
3411 res
+= '%-5s' % fdict
['acodec']
3412 elif fdict
.get('abr') is not None:
3416 if fdict
.get('abr') is not None:
3417 res
+= '@%3dk' % fdict
['abr']
3418 if fdict
.get('asr') is not None:
3419 res
+= ' (%5dHz)' % fdict
['asr']
3420 if fdict
.get('filesize') is not None:
3423 res
+= format_bytes(fdict
['filesize'])
3424 elif fdict
.get('filesize_approx') is not None:
3427 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3430 def render_formats_table(self
, info_dict
):
3431 if not info_dict
.get('formats') and not info_dict
.get('url'):
3434 formats
= info_dict
.get('formats', [info_dict
])
3435 if not self
.params
.get('listformats_table', True) is not False:
3438 format_field(f
, 'format_id'),
3439 format_field(f
, 'ext'),
3440 self
.format_resolution(f
),
3441 self
._format
_note
(f
)
3442 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3443 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3445 delim
= self
._format
_screen
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3448 self
._format
_screen
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3449 format_field(f
, 'ext'),
3450 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3451 format_field(f
, 'fps', '\t%d'),
3452 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3454 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3455 format_field(f
, 'tbr', '\t%dk'),
3456 shorten_protocol_name(f
.get('protocol', '')),
3458 format_field(f
, 'vcodec', default
='unknown').replace(
3459 'none', 'images' if f
.get('acodec') == 'none'
3460 else self
._format
_screen
('audio only', self
.Styles
.SUPPRESS
)),
3461 format_field(f
, 'vbr', '\t%dk'),
3462 format_field(f
, 'acodec', default
='unknown').replace(
3463 'none', '' if f
.get('vcodec') == 'none'
3464 else self
._format
_screen
('video only', self
.Styles
.SUPPRESS
)),
3465 format_field(f
, 'abr', '\t%dk'),
3466 format_field(f
, 'asr', '\t%dHz'),
3468 self
._format
_screen
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3469 format_field(f
, 'language', '[%s]'),
3470 join_nonempty(format_field(f
, 'format_note'),
3471 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3474 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3475 header_line
= self
._list
_format
_headers
(
3476 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3477 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3479 return render_table(
3480 header_line
, table
, hide_empty
=True,
3481 delim
=self
._format
_screen
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3483 def render_thumbnails_table(self
, info_dict
):
3484 thumbnails
= list(info_dict
.get('thumbnails'))
3487 return render_table(
3488 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3489 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3491 def render_subtitles_table(self
, video_id
, subtitles
):
3492 def _row(lang
, formats
):
3493 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3494 if len(set(names
)) == 1:
3495 names
= [] if names
[0] == 'unknown' else names
[:1]
3496 return [lang
, ', '.join(names
), ', '.join(exts
)]
3500 return render_table(
3501 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3502 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3505 def __list_table(self
, video_id
, name
, func
, *args
):
3508 self
.to_screen(f
'{video_id} has no {name}')
3510 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3511 self
.to_stdout(table
)
3513 def list_formats(self
, info_dict
):
3514 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3516 def list_thumbnails(self
, info_dict
):
3517 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3519 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3520 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3522 def urlopen(self
, req
):
3523 """ Start an HTTP download """
3524 if isinstance(req
, compat_basestring
):
3525 req
= sanitized_Request(req
)
3526 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3528 def print_debug_header(self
):
3529 if not self
.params
.get('verbose'):
3532 def get_encoding(stream
):
3533 ret
= getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__)
3534 if not supports_terminal_sequences(stream
):
3535 from .compat
import WINDOWS_VT_MODE
3536 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3539 encoding_str
= 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3540 locale
.getpreferredencoding(),
3541 sys
.getfilesystemencoding(),
3542 get_encoding(self
._screen
_file
), get_encoding(self
._err
_file
),
3543 self
.get_encoding())
3545 logger
= self
.params
.get('logger')
3547 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3548 write_debug(encoding_str
)
3550 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3551 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3553 source
= detect_variant()
3554 write_debug(join_nonempty(
3555 'yt-dlp version', __version__
,
3556 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3557 '' if source
== 'unknown' else f
'({source})',
3559 if not _LAZY_LOADER
:
3560 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3561 write_debug('Lazy loading extractors is forcibly disabled')
3563 write_debug('Lazy loading extractors is disabled')
3564 if plugin_extractors
or plugin_postprocessors
:
3565 write_debug('Plugins: %s' % [
3566 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3567 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3568 if self
.params
.get('compat_opts'):
3569 write_debug('Compatibility options: %s' % ', '.join(self
.params
.get('compat_opts')))
3571 if source
== 'source':
3574 ['git', 'rev-parse', '--short', 'HEAD'],
3575 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3576 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3577 out
, err
= sp
.communicate_or_kill()
3578 out
= out
.decode().strip()
3579 if re
.match('[0-9a-f]+', out
):
3580 write_debug('Git HEAD: %s' % out
)
3587 def python_implementation():
3588 impl_name
= platform
.python_implementation()
3589 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3590 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3593 write_debug('Python version %s (%s %s) - %s' % (
3594 platform
.python_version(),
3595 python_implementation(),
3596 platform
.architecture()[0],
3599 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3600 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3602 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features
)
3604 exe_versions
['rtmpdump'] = rtmpdump_version()
3605 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3606 exe_str
= ', '.join(
3607 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3609 write_debug('exe versions: %s' % exe_str
)
3611 from .downloader
.websocket
import has_websockets
3612 from .postprocessor
.embedthumbnail
import has_mutagen
3613 from .cookies
import SQLITE_AVAILABLE
, SECRETSTORAGE_AVAILABLE
3615 lib_str
= join_nonempty(
3616 compat_pycrypto_AES
and compat_pycrypto_AES
.__name
__.split('.')[0],
3617 SECRETSTORAGE_AVAILABLE
and 'secretstorage',
3618 has_mutagen
and 'mutagen',
3619 SQLITE_AVAILABLE
and 'sqlite',
3620 has_websockets
and 'websockets',
3621 delim
=', ') or 'none'
3622 write_debug('Optional libraries: %s' % lib_str
)
3625 for handler
in self
._opener
.handlers
:
3626 if hasattr(handler
, 'proxies'):
3627 proxy_map
.update(handler
.proxies
)
3628 write_debug(f
'Proxy map: {proxy_map}')
3631 if False and self
.params
.get('call_home'):
3632 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3633 write_debug('Public IP address: %s' % ipaddr
)
3634 latest_version
= self
.urlopen(
3635 'https://yt-dl.org/latest/version').read().decode('utf-8')
3636 if version_tuple(latest_version
) > version_tuple(__version__
):
3637 self
.report_warning(
3638 'You are using an outdated version (newest version: %s)! '
3639 'See https://yt-dl.org/update if you need help updating.' %
3642 def _setup_opener(self
):
3643 timeout_val
= self
.params
.get('socket_timeout')
3644 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3646 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3647 opts_cookiefile
= self
.params
.get('cookiefile')
3648 opts_proxy
= self
.params
.get('proxy')
3650 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3652 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3653 if opts_proxy
is not None:
3654 if opts_proxy
== '':
3657 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3659 proxies
= compat_urllib_request
.getproxies()
3660 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3661 if 'http' in proxies
and 'https' not in proxies
:
3662 proxies
['https'] = proxies
['http']
3663 proxy_handler
= PerRequestProxyHandler(proxies
)
3665 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3666 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3667 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3668 redirect_handler
= YoutubeDLRedirectHandler()
3669 data_handler
= compat_urllib_request_DataHandler()
3671 # When passing our own FileHandler instance, build_opener won't add the
3672 # default FileHandler and allows us to disable the file protocol, which
3673 # can be used for malicious purposes (see
3674 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3675 file_handler
= compat_urllib_request
.FileHandler()
3677 def file_open(*args
, **kwargs
):
3678 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3679 file_handler
.file_open
= file_open
3681 opener
= compat_urllib_request
.build_opener(
3682 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3684 # Delete the default user-agent header, which would otherwise apply in
3685 # cases where our custom HTTP handler doesn't come into play
3686 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3687 opener
.addheaders
= []
3688 self
._opener
= opener
3690 def encode(self
, s
):
3691 if isinstance(s
, bytes):
3692 return s
# Already encoded
3695 return s
.encode(self
.get_encoding())
3696 except UnicodeEncodeError as err
:
3697 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3700 def get_encoding(self
):
3701 encoding
= self
.params
.get('encoding')
3702 if encoding
is None:
3703 encoding
= preferredencoding()
3706 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3707 ''' Write infojson and returns True = written, False = skip, None = error '''
3708 if overwrite
is None:
3709 overwrite
= self
.params
.get('overwrites', True)
3710 if not self
.params
.get('writeinfojson'):
3713 self
.write_debug(f
'Skipping writing {label} infojson')
3715 elif not self
._ensure
_dir
_exists
(infofn
):
3717 elif not overwrite
and os
.path
.exists(infofn
):
3718 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3720 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3722 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3723 except (OSError, IOError):
3724 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3728 def _write_description(self
, label
, ie_result
, descfn
):
3729 ''' Write description and returns True = written, False = skip, None = error '''
3730 if not self
.params
.get('writedescription'):
3733 self
.write_debug(f
'Skipping writing {label} description')
3735 elif not self
._ensure
_dir
_exists
(descfn
):
3737 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3738 self
.to_screen(f
'[info] {label.title()} description is already present')
3739 elif ie_result
.get('description') is None:
3740 self
.report_warning(f
'There\'s no {label} description to write')
3744 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3745 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3746 descfile
.write(ie_result
['description'])
3747 except (OSError, IOError):
3748 self
.report_error(f
'Cannot write {label} description file {descfn}')
3752 def _write_subtitles(self
, info_dict
, filename
):
3753 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3755 subtitles
= info_dict
.get('requested_subtitles')
3756 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3757 # subtitles download errors are already managed as troubles in relevant IE
3758 # that way it will silently go on when used with unsupporting IE
3761 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3762 if not sub_filename_base
:
3763 self
.to_screen('[info] Skipping writing video subtitles')
3765 for sub_lang
, sub_info
in subtitles
.items():
3766 sub_format
= sub_info
['ext']
3767 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3768 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3769 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3771 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3772 sub_info
['filepath'] = existing_sub
3773 ret
.append((existing_sub
, sub_filename_final
))
3776 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3777 if sub_info
.get('data') is not None:
3779 # Use newline='' to prevent conversion of newline characters
3780 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3781 with io
.open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3782 subfile
.write(sub_info
['data'])
3783 sub_info
['filepath'] = sub_filename
3784 ret
.append((sub_filename
, sub_filename_final
))
3786 except (OSError, IOError):
3787 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3791 sub_copy
= sub_info
.copy()
3792 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3793 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3794 sub_info
['filepath'] = sub_filename
3795 ret
.append((sub_filename
, sub_filename_final
))
3796 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3797 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3798 raise DownloadError(f
'Unable to download video subtitles for {sub_lang!r}: {err}', err
)
3799 self
.report_warning(f
'Unable to download video subtitles for {sub_lang!r}: {err}')
3802 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3803 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3804 write_all
= self
.params
.get('write_all_thumbnails', False)
3805 thumbnails
, ret
= [], []
3806 if write_all
or self
.params
.get('writethumbnail', False):
3807 thumbnails
= info_dict
.get('thumbnails') or []
3808 multiple
= write_all
and len(thumbnails
) > 1
3810 if thumb_filename_base
is None:
3811 thumb_filename_base
= filename
3812 if thumbnails
and not thumb_filename_base
:
3813 self
.write_debug(f
'Skipping writing {label} thumbnail')
3816 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3817 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3818 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3819 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3820 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3822 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3824 self
.to_screen('[info] %s is already present' % (
3825 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3826 t
['filepath'] = existing_thumb
3827 ret
.append((existing_thumb
, thumb_filename_final
))
3829 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3831 uf
= self
.urlopen(t
['url'])
3832 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3833 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3834 shutil
.copyfileobj(uf
, thumbf
)
3835 ret
.append((thumb_filename
, thumb_filename_final
))
3836 t
['filepath'] = thumb_filename
3837 except network_exceptions
as err
:
3839 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3840 if ret
and not write_all
: