4 from __future__
import absolute_import
, unicode_literals
32 from string
import ascii_letters
36 compat_get_terminal_size
,
43 compat_tokenize_tokenize
,
45 compat_urllib_request
,
46 compat_urllib_request_DataHandler
,
47 windows_enable_vt_mode
,
49 from .cookies
import load_cookies
90 PerRequestProxyHandler
,
96 register_socks_protocols
,
111 supports_terminal_sequences
,
116 UnavailableVideoError
,
122 YoutubeDLCookieProcessor
,
124 YoutubeDLRedirectHandler
,
126 from .cache
import Cache
127 from .minicurses
import format_text
128 from .extractor
import (
129 gen_extractor_classes
,
132 _PLUGIN_CLASSES
as plugin_extractors
134 from .extractor
.openload
import PhantomJSwrapper
135 from .downloader
import (
137 get_suitable_downloader
,
138 shorten_protocol_name
140 from .downloader
.rtmp
import rtmpdump_version
141 from .postprocessor
import (
144 FFmpegFixupDurationPP
,
147 FFmpegFixupStretchedPP
,
148 FFmpegFixupTimestampPP
,
151 MoveFilesAfterDownloadPP
,
152 _PLUGIN_CLASSES
as plugin_postprocessors
154 from .update
import detect_variant
155 from .version
import __version__
157 if compat_os_name
== 'nt':
161 class YoutubeDL(object):
164 YoutubeDL objects are the ones responsible of downloading the
165 actual video file and writing it to disk if the user has requested
166 it, among some other tasks. In most cases there should be one per
167 program. As, given a video URL, the downloader doesn't know how to
168 extract all the needed information, task that InfoExtractors do, it
169 has to pass the URL to one of them.
171 For this, YoutubeDL objects have a method that allows
172 InfoExtractors to be registered in a given order. When it is passed
173 a URL, the YoutubeDL object handles it to the first InfoExtractor it
174 finds that reports being able to handle it. The InfoExtractor extracts
175 all the information about the video or videos the URL refers to, and
176 YoutubeDL process the extracted information, possibly using a File
177 Downloader to download the video.
179 YoutubeDL objects accept a lot of parameters. In order not to saturate
180 the object constructor with arguments, it receives a dictionary of
181 options instead. These options are available through the params
182 attribute for the InfoExtractors to use. The YoutubeDL also
183 registers itself as the downloader in charge for the InfoExtractors
184 that are added to it, so this is a "mutual registration".
188 username: Username for authentication purposes.
189 password: Password for authentication purposes.
190 videopassword: Password for accessing a video.
191 ap_mso: Adobe Pass multiple-system operator identifier.
192 ap_username: Multiple-system operator account username.
193 ap_password: Multiple-system operator account password.
194 usenetrc: Use netrc for authentication instead.
195 verbose: Print additional info to stdout.
196 quiet: Do not print messages to stdout.
197 no_warnings: Do not print out anything for warnings.
198 forceprint: A list of templates to force print
199 forceurl: Force printing final URL. (Deprecated)
200 forcetitle: Force printing title. (Deprecated)
201 forceid: Force printing ID. (Deprecated)
202 forcethumbnail: Force printing thumbnail URL. (Deprecated)
203 forcedescription: Force printing description. (Deprecated)
204 forcefilename: Force printing final filename. (Deprecated)
205 forceduration: Force printing duration. (Deprecated)
206 forcejson: Force printing info_dict as JSON.
207 dump_single_json: Force printing the info_dict of the whole playlist
208 (or video) as a single JSON line.
209 force_write_download_archive: Force writing download archive regardless
210 of 'skip_download' or 'simulate'.
211 simulate: Do not download the video files. If unset (or None),
212 simulate only if listsubtitles, listformats or list_thumbnails is used
213 format: Video format code. see "FORMAT SELECTION" for more details.
214 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
215 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
216 extracting metadata even if the video is not actually
217 available for download (experimental)
218 format_sort: A list of fields by which to sort the video formats.
219 See "Sorting Formats" for more details.
220 format_sort_force: Force the given format_sort. see "Sorting Formats"
222 allow_multiple_video_streams: Allow multiple video streams to be merged
224 allow_multiple_audio_streams: Allow multiple audio streams to be merged
226 check_formats Whether to test if the formats are downloadable.
227 Can be True (check all), False (check none),
228 'selected' (check selected formats),
229 or None (check only if requested by extractor)
230 paths: Dictionary of output paths. The allowed keys are 'home'
231 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
232 outtmpl: Dictionary of templates for output names. Allowed keys
233 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
234 For compatibility with youtube-dl, a single string can also be used
235 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
236 restrictfilenames: Do not allow "&" and spaces in file names
237 trim_file_name: Limit length of filename (extension excluded)
238 windowsfilenames: Force the filenames to be windows compatible
239 ignoreerrors: Do not stop on download/postprocessing errors.
240 Can be 'only_download' to ignore only download errors.
241 Default is 'only_download' for CLI, but False for API
242 skip_playlist_after_errors: Number of allowed failures until the rest of
243 the playlist is skipped
244 force_generic_extractor: Force downloader to use the generic extractor
245 overwrites: Overwrite all video and metadata files if True,
246 overwrite only non-video files if None
247 and don't overwrite any file if False
248 For compatibility with youtube-dl,
249 "nooverwrites" may also be used instead
250 playliststart: Playlist item to start at.
251 playlistend: Playlist item to end at.
252 playlist_items: Specific indices of playlist to download.
253 playlistreverse: Download playlist items in reverse order.
254 playlistrandom: Download playlist items in random order.
255 matchtitle: Download only matching titles.
256 rejecttitle: Reject downloads for matching titles.
257 logger: Log messages to a logging.Logger instance.
258 logtostderr: Log messages to stderr instead of stdout.
259 consoletitle: Display progress in console window's titlebar.
260 writedescription: Write the video description to a .description file
261 writeinfojson: Write the video description to a .info.json file
262 clean_infojson: Remove private fields from the infojson
263 getcomments: Extract video comments. This will not be written to disk
264 unless writeinfojson is also given
265 writeannotations: Write the video annotations to a .annotations.xml file
266 writethumbnail: Write the thumbnail image to a file
267 allow_playlist_files: Whether to write playlists' description, infojson etc
268 also to disk when using the 'write*' options
269 write_all_thumbnails: Write all thumbnail formats to files
270 writelink: Write an internet shortcut file, depending on the
271 current platform (.url/.webloc/.desktop)
272 writeurllink: Write a Windows internet shortcut file (.url)
273 writewebloclink: Write a macOS internet shortcut file (.webloc)
274 writedesktoplink: Write a Linux internet shortcut file (.desktop)
275 writesubtitles: Write the video subtitles to a file
276 writeautomaticsub: Write the automatically generated subtitles to a file
277 allsubtitles: Deprecated - Use subtitleslangs = ['all']
278 Downloads all the subtitles of the video
279 (requires writesubtitles or writeautomaticsub)
280 listsubtitles: Lists all available subtitles for the video
281 subtitlesformat: The format code for subtitles
282 subtitleslangs: List of languages of the subtitles to download (can be regex).
283 The list may contain "all" to refer to all the available
284 subtitles. The language can be prefixed with a "-" to
285 exclude it from the requested languages. Eg: ['all', '-live_chat']
286 keepvideo: Keep the video file after post-processing
287 daterange: A DateRange object, download only if the upload_date is in the range.
288 skip_download: Skip the actual download of the video file
289 cachedir: Location of the cache files in the filesystem.
290 False to disable filesystem cache.
291 noplaylist: Download single video instead of a playlist if in doubt.
292 age_limit: An integer representing the user's age in years.
293 Unsuitable videos for the given age are skipped.
294 min_views: An integer representing the minimum view count the video
295 must have in order to not be skipped.
296 Videos without view count information are always
297 downloaded. None for no limit.
298 max_views: An integer representing the maximum view count.
299 Videos that are more popular than that are not
301 Videos without view count information are always
302 downloaded. None for no limit.
303 download_archive: File name of a file where all downloads are recorded.
304 Videos already present in the file are not downloaded
306 break_on_existing: Stop the download process after attempting to download a
307 file that is in the archive.
308 break_on_reject: Stop the download process when encountering a video that
309 has been filtered out.
310 cookiefile: File name where cookies should be read from and dumped to
311 cookiesfrombrowser: A tuple containing the name of the browser and the profile
312 name/path from where cookies are loaded.
313 Eg: ('chrome', ) or ('vivaldi', 'default')
314 nocheckcertificate:Do not verify SSL certificates
315 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
316 At the moment, this is only supported by YouTube.
317 proxy: URL of the proxy server to use
318 geo_verification_proxy: URL of the proxy to use for IP address verification
319 on geo-restricted sites.
320 socket_timeout: Time to wait for unresponsive hosts, in seconds
321 bidi_workaround: Work around buggy terminals without bidirectional text
322 support, using fridibi
323 debug_printtraffic:Print out sent and received HTTP traffic
324 include_ads: Download ads as well
325 default_search: Prepend this string if an input url is not valid.
326 'auto' for elaborate guessing
327 encoding: Use this encoding instead of the system-specified.
328 extract_flat: Do not resolve URLs, return the immediate result.
329 Pass in 'in_playlist' to only show this behavior for
331 postprocessors: A list of dictionaries, each with an entry
332 * key: The name of the postprocessor. See
333 yt_dlp/postprocessor/__init__.py for a list.
334 * when: When to run the postprocessor. Can be one of
335 pre_process|before_dl|post_process|after_move.
336 Assumed to be 'post_process' if not given
337 post_hooks: Deprecated - Register a custom postprocessor instead
338 A list of functions that get called as the final step
339 for each video file, after all postprocessors have been
340 called. The filename will be passed as the only argument.
341 progress_hooks: A list of functions that get called on download
342 progress, with a dictionary with the entries
343 * status: One of "downloading", "error", or "finished".
344 Check this first and ignore unknown values.
345 * info_dict: The extracted info_dict
347 If status is one of "downloading", or "finished", the
348 following properties may also be present:
349 * filename: The final filename (always present)
350 * tmpfilename: The filename we're currently writing to
351 * downloaded_bytes: Bytes on disk
352 * total_bytes: Size of the whole file, None if unknown
353 * total_bytes_estimate: Guess of the eventual file size,
355 * elapsed: The number of seconds since download started.
356 * eta: The estimated time in seconds, None if unknown
357 * speed: The download speed in bytes/second, None if
359 * fragment_index: The counter of the currently
360 downloaded video fragment.
361 * fragment_count: The number of fragments (= individual
362 files that will be merged)
364 Progress hooks are guaranteed to be called at least once
365 (with status "finished") if the download is successful.
366 postprocessor_hooks: A list of functions that get called on postprocessing
367 progress, with a dictionary with the entries
368 * status: One of "started", "processing", or "finished".
369 Check this first and ignore unknown values.
370 * postprocessor: Name of the postprocessor
371 * info_dict: The extracted info_dict
373 Progress hooks are guaranteed to be called at least twice
374 (with status "started" and "finished") if the processing is successful.
375 merge_output_format: Extension to use when merging formats.
376 final_ext: Expected final extension; used to detect when the file was
377 already downloaded and converted
378 fixup: Automatically correct known faults of the file.
380 - "never": do nothing
381 - "warn": only emit a warning
382 - "detect_or_warn": check whether we can do anything
383 about it, warn otherwise (default)
384 source_address: Client-side IP address to bind to.
385 call_home: Boolean, true iff we are allowed to contact the
386 yt-dlp servers for debugging. (BROKEN)
387 sleep_interval_requests: Number of seconds to sleep between requests
389 sleep_interval: Number of seconds to sleep before each download when
390 used alone or a lower bound of a range for randomized
391 sleep before each download (minimum possible number
392 of seconds to sleep) when used along with
394 max_sleep_interval:Upper bound of a range for randomized sleep before each
395 download (maximum possible number of seconds to sleep).
396 Must only be used along with sleep_interval.
397 Actual sleep time will be a random float from range
398 [sleep_interval; max_sleep_interval].
399 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
400 listformats: Print an overview of available video formats and exit.
401 list_thumbnails: Print a table of all thumbnails and exit.
402 match_filter: A function that gets called with the info_dict of
404 If it returns a message, the video is ignored.
405 If it returns None, the video is downloaded.
406 match_filter_func in utils.py is one example for this.
407 no_color: Do not emit color codes in output.
408 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
411 Two-letter ISO 3166-2 country code that will be used for
412 explicit geographic restriction bypassing via faking
413 X-Forwarded-For HTTP header
415 IP range in CIDR notation that will be used similarly to
418 The following options determine which downloader is picked:
419 external_downloader: A dictionary of protocol keys and the executable of the
420 external downloader to use for it. The allowed protocols
421 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
422 Set the value to 'native' to use the native downloader
423 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
424 or {'m3u8': 'ffmpeg'} instead.
425 Use the native HLS downloader instead of ffmpeg/avconv
426 if True, otherwise use ffmpeg/avconv if False, otherwise
427 use downloader suggested by extractor if None.
428 compat_opts: Compatibility options. See "Differences in default behavior".
429 The following options do not work when used through the API:
430 filename, abort-on-error, multistreams, no-live-chat, format-sort
431 no-clean-infojson, no-playlist-metafiles, no-keep-subs.
432 Refer __init__.py for their implementation
433 progress_template: Dictionary of templates for progress outputs.
434 Allowed keys are 'download', 'postprocess',
435 'download-title' (console title) and 'postprocess-title'.
436 The template is mapped on a dictionary with keys 'progress' and 'info'
438 The following parameters are not used by YoutubeDL itself, they are used by
439 the downloader (see yt_dlp/downloader/common.py):
440 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
441 max_filesize, test, noresizebuffer, retries, fragment_retries, continuedl,
442 noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
443 external_downloader_args, concurrent_fragment_downloads.
445 The following options are used by the post processors:
446 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
447 otherwise prefer ffmpeg. (avconv support is deprecated)
448 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
449 to the binary or its containing directory.
450 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
451 and a list of additional command-line arguments for the
452 postprocessor/executable. The dict can also have "PP+EXE" keys
453 which are used when the given exe is used by the given PP.
454 Use 'default' as the name for arguments to passed to all PP
455 For compatibility with youtube-dl, a single list of args
458 The following options are used by the extractors:
459 extractor_retries: Number of times to retry for known errors
460 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
461 hls_split_discontinuity: Split HLS playlists to different formats at
462 discontinuities such as ad breaks (default: False)
463 extractor_args: A dictionary of arguments to be passed to the extractors.
464 See "EXTRACTOR ARGUMENTS" for details.
465 Eg: {'youtube': {'skip': ['dash', 'hls']}}
466 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
467 If True (default), DASH manifests and related
468 data will be downloaded and processed by extractor.
469 You can reduce network I/O by disabling it if you don't
470 care about DASH. (only for youtube)
471 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
472 If True (default), HLS manifests and related
473 data will be downloaded and processed by extractor.
474 You can reduce network I/O by disabling it if you don't
475 care about HLS. (only for youtube)
478 _NUMERIC_FIELDS
= set((
479 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
480 'timestamp', 'release_timestamp',
481 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
482 'average_rating', 'comment_count', 'age_limit',
483 'start_time', 'end_time',
484 'chapter_number', 'season_number', 'episode_number',
485 'track_number', 'disc_number', 'release_year',
488 _format_selection_exts
= {
489 'audio': {'m4a', 'mp3', 'ogg', 'aac'}
,
490 'video': {'mp4', 'flv', 'webm', '3gp'}
,
491 'storyboards': {'mhtml'}
,
496 _pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
497 _printed_messages
= set()
498 _first_webpage_request
= True
499 _download_retcode
= None
500 _num_downloads
= None
502 _playlist_urls
= set()
505 def __init__(self
, params
=None, auto_init
=True):
506 """Create a FileDownloader object with the given options.
507 @param auto_init Whether to load the default extractors and print header (if verbose).
508 Set to 'no_verbose_header' to not print the header
513 self
._ies
_instances
= {}
514 self
._pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
515 self
._printed
_messages
= set()
516 self
._first
_webpage
_request
= True
517 self
._post
_hooks
= []
518 self
._progress
_hooks
= []
519 self
._postprocessor
_hooks
= []
520 self
._download
_retcode
= 0
521 self
._num
_downloads
= 0
522 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
523 self
._err
_file
= sys
.stderr
525 self
.cache
= Cache(self
)
527 windows_enable_vt_mode()
528 # FIXME: This will break if we ever print color to stdout
529 self
._allow
_colors
= {
530 'screen': not self
.params
.get('no_color') and supports_terminal_sequences(self
._screen
_file
),
531 'err': not self
.params
.get('no_color') and supports_terminal_sequences(self
._err
_file
),
534 if sys
.version_info
< (3, 6):
536 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
538 if self
.params
.get('allow_unplayable_formats'):
540 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
541 'This is a developer option intended for debugging. \n'
542 ' If you experience any issues while using this option, '
543 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
545 def check_deprecated(param
, option
, suggestion
):
546 if self
.params
.get(param
) is not None:
547 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
551 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
552 if self
.params
.get('geo_verification_proxy') is None:
553 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
555 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
556 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
557 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
559 for msg
in self
.params
.get('_warnings', []):
560 self
.report_warning(msg
)
562 if 'list-formats' in self
.params
.get('compat_opts', []):
563 self
.params
['listformats_table'] = False
565 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
566 # nooverwrites was unnecessarily changed to overwrites
567 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
568 # This ensures compatibility with both keys
569 self
.params
['overwrites'] = not self
.params
['nooverwrites']
570 elif self
.params
.get('overwrites') is None:
571 self
.params
.pop('overwrites', None)
573 self
.params
['nooverwrites'] = not self
.params
['overwrites']
575 if params
.get('bidi_workaround', False):
578 master
, slave
= pty
.openpty()
579 width
= compat_get_terminal_size().columns
583 width_args
= ['-w', str(width
)]
585 stdin
=subprocess
.PIPE
,
587 stderr
=self
._err
_file
)
589 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
591 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
592 self
._output
_channel
= os
.fdopen(master
, 'rb')
593 except OSError as ose
:
594 if ose
.errno
== errno
.ENOENT
:
596 'Could not find fribidi executable, ignoring --bidi-workaround. '
597 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
601 if (sys
.platform
!= 'win32'
602 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
603 and not params
.get('restrictfilenames', False)):
604 # Unicode filesystem API will throw errors (#1474, #13027)
606 'Assuming --restrict-filenames since file system encoding '
607 'cannot encode all characters. '
608 'Set the LC_ALL environment variable to fix this.')
609 self
.params
['restrictfilenames'] = True
611 self
.outtmpl_dict
= self
.parse_outtmpl()
613 # Creating format selector here allows us to catch syntax errors before the extraction
614 self
.format_selector
= (
615 None if self
.params
.get('format') is None
616 else self
.build_format_selector(self
.params
['format']))
621 if auto_init
!= 'no_verbose_header':
622 self
.print_debug_header()
623 self
.add_default_info_extractors()
625 for pp_def_raw
in self
.params
.get('postprocessors', []):
626 pp_def
= dict(pp_def_raw
)
627 when
= pp_def
.pop('when', 'post_process')
628 pp_class
= get_postprocessor(pp_def
.pop('key'))
629 pp
= pp_class(self
, **compat_kwargs(pp_def
))
630 self
.add_post_processor(pp
, when
=when
)
632 for ph
in self
.params
.get('post_hooks', []):
633 self
.add_post_hook(ph
)
635 for ph
in self
.params
.get('progress_hooks', []):
636 self
.add_progress_hook(ph
)
638 register_socks_protocols()
640 def preload_download_archive(fn
):
641 """Preload the archive, if any is specified"""
644 self
.write_debug(f
'Loading archive file {fn!r}')
646 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
647 for line
in archive_file
:
648 self
.archive
.add(line
.strip())
649 except IOError as ioe
:
650 if ioe
.errno
!= errno
.ENOENT
:
656 preload_download_archive(self
.params
.get('download_archive'))
658 def warn_if_short_id(self
, argv
):
659 # short YouTube ID starting with dash?
661 i
for i
, a
in enumerate(argv
)
662 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
666 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
667 + ['--'] + [argv
[i
] for i
in idxs
]
670 'Long argument string detected. '
671 'Use -- to separate parameters and URLs, like this:\n%s' %
672 args_to_str(correct_argv
))
674 def add_info_extractor(self
, ie
):
675 """Add an InfoExtractor object to the end of the list."""
677 self
._ies
[ie_key
] = ie
678 if not isinstance(ie
, type):
679 self
._ies
_instances
[ie_key
] = ie
680 ie
.set_downloader(self
)
682 def _get_info_extractor_class(self
, ie_key
):
683 ie
= self
._ies
.get(ie_key
)
685 ie
= get_info_extractor(ie_key
)
686 self
.add_info_extractor(ie
)
689 def get_info_extractor(self
, ie_key
):
691 Get an instance of an IE with name ie_key, it will try to get one from
692 the _ies list, if there's no instance it will create a new one and add
693 it to the extractor list.
695 ie
= self
._ies
_instances
.get(ie_key
)
697 ie
= get_info_extractor(ie_key
)()
698 self
.add_info_extractor(ie
)
701 def add_default_info_extractors(self
):
703 Add the InfoExtractors returned by gen_extractors to the end of the list
705 for ie
in gen_extractor_classes():
706 self
.add_info_extractor(ie
)
708 def add_post_processor(self
, pp
, when
='post_process'):
709 """Add a PostProcessor object to the end of the chain."""
710 self
._pps
[when
].append(pp
)
711 pp
.set_downloader(self
)
713 def add_post_hook(self
, ph
):
714 """Add the post hook"""
715 self
._post
_hooks
.append(ph
)
717 def add_progress_hook(self
, ph
):
718 """Add the download progress hook"""
719 self
._progress
_hooks
.append(ph
)
721 def add_postprocessor_hook(self
, ph
):
722 """Add the postprocessing progress hook"""
723 self
._postprocessor
_hooks
.append(ph
)
725 def _bidi_workaround(self
, message
):
726 if not hasattr(self
, '_output_channel'):
729 assert hasattr(self
, '_output_process')
730 assert isinstance(message
, compat_str
)
731 line_count
= message
.count('\n') + 1
732 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
733 self
._output
_process
.stdin
.flush()
734 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
735 for _
in range(line_count
))
736 return res
[:-len('\n')]
738 def _write_string(self
, message
, out
=None, only_once
=False):
740 if message
in self
._printed
_messages
:
742 self
._printed
_messages
.add(message
)
743 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
745 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
746 """Print message to stdout"""
747 if self
.params
.get('logger'):
748 self
.params
['logger'].debug(message
)
749 elif not quiet
or self
.params
.get('verbose'):
751 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
752 self
._err
_file
if quiet
else self
._screen
_file
)
754 def to_stderr(self
, message
, only_once
=False):
755 """Print message to stderr"""
756 assert isinstance(message
, compat_str
)
757 if self
.params
.get('logger'):
758 self
.params
['logger'].error(message
)
760 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._err
_file
, only_once
=only_once
)
762 def to_console_title(self
, message
):
763 if not self
.params
.get('consoletitle', False):
765 if compat_os_name
== 'nt':
766 if ctypes
.windll
.kernel32
.GetConsoleWindow():
767 # c_wchar_p() might not be necessary if `message` is
768 # already of type unicode()
769 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
770 elif 'TERM' in os
.environ
:
771 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
773 def save_console_title(self
):
774 if not self
.params
.get('consoletitle', False):
776 if self
.params
.get('simulate'):
778 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
779 # Save the title on stack
780 self
._write
_string
('\033[22;0t', self
._screen
_file
)
782 def restore_console_title(self
):
783 if not self
.params
.get('consoletitle', False):
785 if self
.params
.get('simulate'):
787 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
788 # Restore the title from stack
789 self
._write
_string
('\033[23;0t', self
._screen
_file
)
792 self
.save_console_title()
795 def __exit__(self
, *args
):
796 self
.restore_console_title()
798 if self
.params
.get('cookiefile') is not None:
799 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
801 def trouble(self
, message
=None, tb
=None):
802 """Determine action to take when a download problem appears.
804 Depending on if the downloader has been configured to ignore
805 download errors or not, this method may throw an exception or
806 not when errors are found, after printing the message.
808 tb, if given, is additional traceback information.
810 if message
is not None:
811 self
.to_stderr(message
)
812 if self
.params
.get('verbose'):
814 if sys
.exc_info()[0]: # if .trouble has been called from an except block
816 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
817 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
818 tb
+= encode_compat_str(traceback
.format_exc())
820 tb_data
= traceback
.format_list(traceback
.extract_stack())
821 tb
= ''.join(tb_data
)
824 if not self
.params
.get('ignoreerrors'):
825 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
826 exc_info
= sys
.exc_info()[1].exc_info
828 exc_info
= sys
.exc_info()
829 raise DownloadError(message
, exc_info
)
830 self
._download
_retcode
= 1
832 def to_screen(self
, message
, skip_eol
=False):
833 """Print message to stdout if not in quiet mode"""
835 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
845 def __format_text(self
, out
, text
, f
, fallback
=None, *, test_encoding
=False):
846 assert out
in ('screen', 'err')
849 handle
= self
._screen
_file
if out
== 'screen' else self
._err
_file
850 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', 'ascii')
851 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
852 if fallback
is not None and text
!= original_text
:
854 if isinstance(f
, self
.Styles
):
856 return format_text(text
, f
) if self
._allow
_colors
[out
] else text
if fallback
is None else fallback
858 def _format_screen(self
, *args
, **kwargs
):
859 return self
.__format
_text
('screen', *args
, **kwargs
)
861 def _format_err(self
, *args
, **kwargs
):
862 return self
.__format
_text
('err', *args
, **kwargs
)
864 def report_warning(self
, message
, only_once
=False):
866 Print the message to stderr, it will be prefixed with 'WARNING:'
867 If stderr is a tty file the 'WARNING:' will be colored
869 if self
.params
.get('logger') is not None:
870 self
.params
['logger'].warning(message
)
872 if self
.params
.get('no_warnings'):
874 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
876 def report_error(self
, message
, tb
=None):
878 Do the same as trouble, but prefixes the message with 'ERROR:', colored
879 in red if stderr is a tty file.
881 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', tb
)
883 def write_debug(self
, message
, only_once
=False):
884 '''Log debug message or Print message to stderr'''
885 if not self
.params
.get('verbose', False):
887 message
= '[debug] %s' % message
888 if self
.params
.get('logger'):
889 self
.params
['logger'].debug(message
)
891 self
.to_stderr(message
, only_once
)
893 def report_file_already_downloaded(self
, file_name
):
894 """Report file has already been fully downloaded."""
896 self
.to_screen('[download] %s has already been downloaded' % file_name
)
897 except UnicodeEncodeError:
898 self
.to_screen('[download] The file has already been downloaded')
900 def report_file_delete(self
, file_name
):
901 """Report that existing file will be deleted."""
903 self
.to_screen('Deleting existing file %s' % file_name
)
904 except UnicodeEncodeError:
905 self
.to_screen('Deleting existing file')
907 def raise_no_formats(self
, info
, forced
=False):
908 has_drm
= info
.get('__has_drm')
909 msg
= 'This video is DRM protected' if has_drm
else 'No video formats found!'
910 expected
= self
.params
.get('ignore_no_formats_error')
911 if forced
or not expected
:
912 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
913 expected
=has_drm
or expected
)
915 self
.report_warning(msg
)
917 def parse_outtmpl(self
):
918 outtmpl_dict
= self
.params
.get('outtmpl', {})
919 if not isinstance(outtmpl_dict
, dict):
920 outtmpl_dict
= {'default': outtmpl_dict}
921 # Remove spaces in the default template
922 if self
.params
.get('restrictfilenames'):
923 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
925 sanitize
= lambda x
: x
926 outtmpl_dict
.update({
927 k
: sanitize(v
) for k
, v
in DEFAULT_OUTTMPL
.items()
928 if outtmpl_dict
.get(k
) is None})
929 for key
, val
in outtmpl_dict
.items():
930 if isinstance(val
, bytes):
932 'Parameter outtmpl is bytes, but should be a unicode string. '
933 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
936 def get_output_path(self
, dir_type
='', filename
=None):
937 paths
= self
.params
.get('paths', {})
938 assert isinstance(paths
, dict)
940 expand_path(paths
.get('home', '').strip()),
941 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
944 # Temporary fix for #4787
945 # 'Treat' all problem characters by passing filename through preferredencoding
946 # to workaround encoding issues with subprocess on python2 @ Windows
947 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
948 path
= encodeFilename(path
, True).decode(preferredencoding())
949 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
952 def _outtmpl_expandpath(outtmpl
):
953 # expand_path translates '%%' into '%' and '$$' into '$'
954 # correspondingly that is not what we want since we need to keep
955 # '%%' intact for template dict substitution step. Working around
956 # with boundary-alike separator hack.
957 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
958 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
960 # outtmpl should be expand_path'ed before template dict substitution
961 # because meta fields may contain env variables we don't want to
962 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
963 # title "Hello $PATH", we don't want `$PATH` to be expanded.
964 return expand_path(outtmpl
).replace(sep
, '')
967 def escape_outtmpl(outtmpl
):
968 ''' Escape any remaining strings like %s, %abc% etc. '''
970 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
971 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
975 def validate_outtmpl(cls
, outtmpl
):
976 ''' @return None or Exception object '''
978 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljqBU]'),
979 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
980 cls
._outtmpl
_expandpath
(outtmpl
))
982 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
984 except ValueError as err
:
988 def _copy_infodict(info_dict
):
989 info_dict
= dict(info_dict
)
990 for key
in ('__original_infodict', '__postprocessors'):
991 info_dict
.pop(key
, None)
994 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=None):
995 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict """
996 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
998 info_dict
= self
._copy
_infodict
(info_dict
)
999 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1000 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1001 if info_dict
.get('duration', None) is not None
1003 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
1004 if info_dict
.get('resolution') is None:
1005 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1007 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1008 # of %(field)s to %(field)0Nd for backward compatibility
1009 field_size_compat_map
= {
1010 'playlist_index': number_of_digits(info_dict
.get('_last_playlist_index') or 0),
1011 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1012 'autonumber': self
.params
.get('autonumber_size') or 5,
1016 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljqBU]'))
1021 # Field is of the form key1.key2...
1022 # where keys (except first) can be string, int or slice
1023 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1024 MATH_FIELD_RE
= r
'''{field}|{num}'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
1025 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1026 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
1029 (?P<maths>(?:{math_op}{math_field})*)
1030 (?:>(?P<strf_format>.+?))?
1031 (?P<alternate>(?<!\\),[^|)]+)?
1032 (?:\|(?P<default>.*?))?
1033 $'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
1035 def _traverse_infodict(k
):
1039 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1041 def get_value(mdict
):
1043 value
= _traverse_infodict(mdict
['fields'])
1046 value
= float_or_none(value
)
1047 if value
is not None:
1050 offset_key
= mdict
['maths']
1052 value
= float_or_none(value
)
1056 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1057 offset_key
).group(0)
1058 offset_key
= offset_key
[len(item
):]
1059 if operator
is None:
1060 operator
= MATH_FUNCTIONS
[item
]
1062 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1063 offset
= float_or_none(item
)
1065 offset
= float_or_none(_traverse_infodict(item
))
1067 value
= operator(value
, multiplier
* offset
)
1068 except (TypeError, ZeroDivisionError):
1071 # Datetime formatting
1072 if mdict
['strf_format']:
1073 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1077 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1079 def _dumpjson_default(obj
):
1080 if isinstance(obj
, (set, LazyList
)):
1082 raise TypeError(f
'Object of type {type(obj).__name__} is not JSON serializable')
1084 def create_key(outer_mobj
):
1085 if not outer_mobj
.group('has_key'):
1086 return outer_mobj
.group(0)
1087 key
= outer_mobj
.group('key')
1088 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1089 initial_field
= mobj
.group('fields').split('.')[-1] if mobj
else ''
1090 value
, default
= None, na
1092 mobj
= mobj
.groupdict()
1093 default
= mobj
['default'] if mobj
['default'] is not None else default
1094 value
= get_value(mobj
)
1095 if value
is None and mobj
['alternate']:
1096 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['alternate'][1:])
1100 fmt
= outer_mobj
.group('format')
1101 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1102 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
1104 value
= default
if value
is None else value
1106 flags
= outer_mobj
.group('conversion') or ''
1107 str_fmt
= f
'{fmt[:-1]}s'
1108 if fmt
[-1] == 'l': # list
1109 delim
= '\n' if '#' in flags
else ', '
1110 value
, fmt
= delim
.join(variadic(value
)), str_fmt
1111 elif fmt
[-1] == 'j': # json
1112 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1113 elif fmt
[-1] == 'q': # quoted
1114 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1115 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1116 elif fmt
[-1] == 'B': # bytes
1117 value
= f
'%{str_fmt}'.encode('utf-8') % str(value
).encode('utf-8')
1118 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1119 elif fmt
[-1] == 'U': # unicode normalized
1120 value
, fmt
= unicodedata
.normalize(
1121 # "+" = compatibility equivalence, "#" = NFD
1122 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1124 elif fmt
[-1] == 'c':
1126 value
= str(value
)[0]
1129 elif fmt
[-1] not in 'rs': # numeric
1130 value
= float_or_none(value
)
1132 value
, fmt
= default
, 's'
1136 # If value is an object, sanitize might convert it to a string
1137 # So we convert it to repr first
1138 value
, fmt
= repr(value
), str_fmt
1139 if fmt
[-1] in 'csr':
1140 value
= sanitize(initial_field
, value
)
1142 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1143 TMPL_DICT
[key
] = value
1144 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1146 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1148 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1149 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1150 return self
.escape_outtmpl(outtmpl
) % info_dict
1152 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
1154 sanitize
= lambda k
, v
: sanitize_filename(
1156 restricted
=self
.params
.get('restrictfilenames'),
1157 is_id
=(k
== 'id' or k
.endswith('_id')))
1158 outtmpl
= self
._outtmpl
_expandpath
(self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default']))
1159 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, sanitize
)
1161 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
1162 if filename
and force_ext
is not None:
1163 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1165 # https://github.com/blackjack4494/youtube-dlc/issues/85
1166 trim_file_name
= self
.params
.get('trim_file_name', False)
1168 fn_groups
= filename
.rsplit('.')
1171 if len(fn_groups
) > 2:
1172 sub_ext
= fn_groups
[-2]
1173 filename
= join_nonempty(fn_groups
[0][:trim_file_name
], sub_ext
, ext
, delim
='.')
1176 except ValueError as err
:
1177 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1180 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
1181 """Generate the output filename."""
1183 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
1184 if not filename
and dir_type
not in ('', 'temp'):
1188 if not self
.params
.get('paths'):
1190 elif filename
== '-':
1191 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1192 elif os
.path
.isabs(filename
):
1193 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1194 if filename
== '-' or not filename
:
1197 return self
.get_output_path(dir_type
, filename
)
1199 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1200 """ Returns None if the file should be downloaded """
1202 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1205 if 'title' in info_dict
:
1206 # This can happen when we're just evaluating the playlist
1207 title
= info_dict
['title']
1208 matchtitle
= self
.params
.get('matchtitle', False)
1210 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1211 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1212 rejecttitle
= self
.params
.get('rejecttitle', False)
1214 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1215 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1216 date
= info_dict
.get('upload_date')
1217 if date
is not None:
1218 dateRange
= self
.params
.get('daterange', DateRange())
1219 if date
not in dateRange
:
1220 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1221 view_count
= info_dict
.get('view_count')
1222 if view_count
is not None:
1223 min_views
= self
.params
.get('min_views')
1224 if min_views
is not None and view_count
< min_views
:
1225 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1226 max_views
= self
.params
.get('max_views')
1227 if max_views
is not None and view_count
> max_views
:
1228 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1229 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1230 return 'Skipping "%s" because it is age restricted' % video_title
1232 match_filter
= self
.params
.get('match_filter')
1233 if match_filter
is not None:
1235 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1237 # For backward compatibility
1238 ret
= None if incomplete
else match_filter(info_dict
)
1243 if self
.in_download_archive(info_dict
):
1244 reason
= '%s has already been recorded in the archive' % video_title
1245 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1247 reason
= check_filter()
1248 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1249 if reason
is not None:
1251 self
.to_screen('[download] ' + reason
)
1252 if self
.params
.get(break_opt
, False):
1257 def add_extra_info(info_dict
, extra_info
):
1258 '''Set the keys from extra_info in info dict if they are missing'''
1259 for key
, value
in extra_info
.items():
1260 info_dict
.setdefault(key
, value
)
1262 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1263 process
=True, force_generic_extractor
=False):
1265 Return a list with a dictionary for each video extracted.
1268 url -- URL to extract
1271 download -- whether to download videos during extraction
1272 ie_key -- extractor key hint
1273 extra_info -- dictionary containing the extra values to add to each result
1274 process -- whether to resolve all unresolved references (URLs, playlist items),
1275 must be True for download to work.
1276 force_generic_extractor -- force using the generic extractor
1279 if extra_info
is None:
1282 if not ie_key
and force_generic_extractor
:
1286 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1290 for ie_key
, ie
in ies
.items():
1291 if not ie
.suitable(url
):
1294 if not ie
.working():
1295 self
.report_warning('The program functionality for this site has been marked as broken, '
1296 'and will probably not work.')
1298 temp_id
= ie
.get_temp_id(url
)
1299 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1300 self
.to_screen("[%s] %s: has already been recorded in archive" % (
1303 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1305 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1307 def __handle_extraction_exceptions(func
):
1308 @functools.wraps(func
)
1309 def wrapper(self
, *args
, **kwargs
):
1311 return func(self
, *args
, **kwargs
)
1312 except GeoRestrictedError
as e
:
1315 msg
+= '\nThis video is available in %s.' % ', '.join(
1316 map(ISO3166Utils
.short2full
, e
.countries
))
1317 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1318 self
.report_error(msg
)
1319 except ExtractorError
as e
: # An error we somewhat expected
1320 self
.report_error(compat_str(e
), e
.format_traceback())
1321 except ThrottledDownload
as e
:
1322 self
.to_stderr('\r')
1323 self
.report_warning(f
'{e}; Re-extracting data')
1324 return wrapper(self
, *args
, **kwargs
)
1325 except (DownloadCancelled
, LazyList
.IndexError):
1327 except Exception as e
:
1328 if self
.params
.get('ignoreerrors'):
1329 self
.report_error(error_to_compat_str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1334 @__handle_extraction_exceptions
1335 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1336 ie_result
= ie
.extract(url
)
1337 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1339 if isinstance(ie_result
, list):
1340 # Backwards compatibility: old IE result format
1342 '_type': 'compat_list',
1343 'entries': ie_result
,
1345 if extra_info
.get('original_url'):
1346 ie_result
.setdefault('original_url', extra_info
['original_url'])
1347 self
.add_default_extra_info(ie_result
, ie
, url
)
1349 return self
.process_ie_result(ie_result
, download
, extra_info
)
1353 def add_default_extra_info(self
, ie_result
, ie
, url
):
1355 self
.add_extra_info(ie_result
, {
1357 'original_url': url
,
1358 'webpage_url_basename': url_basename(url
),
1361 self
.add_extra_info(ie_result
, {
1362 'extractor': ie
.IE_NAME
,
1363 'extractor_key': ie
.ie_key(),
1366 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1368 Take the result of the ie(may be modified) and resolve all unresolved
1369 references (URLs, playlist items).
1371 It will also download the videos if 'download'.
1372 Returns the resolved ie_result.
1374 if extra_info
is None:
1376 result_type
= ie_result
.get('_type', 'video')
1378 if result_type
in ('url', 'url_transparent'):
1379 ie_result
['url'] = sanitize_url(ie_result
['url'])
1380 if ie_result
.get('original_url'):
1381 extra_info
.setdefault('original_url', ie_result
['original_url'])
1383 extract_flat
= self
.params
.get('extract_flat', False)
1384 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1385 or extract_flat
is True):
1386 info_copy
= ie_result
.copy()
1387 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1388 if ie
and not ie_result
.get('id'):
1389 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1390 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1391 self
.add_extra_info(info_copy
, extra_info
)
1392 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1393 if self
.params
.get('force_write_download_archive', False):
1394 self
.record_download_archive(info_copy
)
1397 if result_type
== 'video':
1398 self
.add_extra_info(ie_result
, extra_info
)
1399 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1400 additional_urls
= (ie_result
or {}).get('additional_urls')
1402 # TODO: Improve MetadataParserPP to allow setting a list
1403 if isinstance(additional_urls
, compat_str
):
1404 additional_urls
= [additional_urls
]
1406 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1407 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1408 ie_result
['additional_entries'] = [
1410 url
, download
, extra_info
,
1411 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1412 for url
in additional_urls
1415 elif result_type
== 'url':
1416 # We have to add extra_info to the results because it may be
1417 # contained in a playlist
1418 return self
.extract_info(
1419 ie_result
['url'], download
,
1420 ie_key
=ie_result
.get('ie_key'),
1421 extra_info
=extra_info
)
1422 elif result_type
== 'url_transparent':
1423 # Use the information from the embedding page
1424 info
= self
.extract_info(
1425 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1426 extra_info
=extra_info
, download
=False, process
=False)
1428 # extract_info may return None when ignoreerrors is enabled and
1429 # extraction failed with an error, don't crash and return early
1434 force_properties
= dict(
1435 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1436 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1437 if f
in force_properties
:
1438 del force_properties
[f
]
1439 new_result
= info
.copy()
1440 new_result
.update(force_properties
)
1442 # Extracted info may not be a video result (i.e.
1443 # info.get('_type', 'video') != video) but rather an url or
1444 # url_transparent. In such cases outer metadata (from ie_result)
1445 # should be propagated to inner one (info). For this to happen
1446 # _type of info should be overridden with url_transparent. This
1447 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1448 if new_result
.get('_type') == 'url':
1449 new_result
['_type'] = 'url_transparent'
1451 return self
.process_ie_result(
1452 new_result
, download
=download
, extra_info
=extra_info
)
1453 elif result_type
in ('playlist', 'multi_video'):
1454 # Protect from infinite recursion due to recursively nested playlists
1455 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1456 webpage_url
= ie_result
['webpage_url']
1457 if webpage_url
in self
._playlist
_urls
:
1459 '[download] Skipping already downloaded playlist: %s'
1460 % ie_result
.get('title') or ie_result
.get('id'))
1463 self
._playlist
_level
+= 1
1464 self
._playlist
_urls
.add(webpage_url
)
1465 self
._sanitize
_thumbnails
(ie_result
)
1467 return self
.__process
_playlist
(ie_result
, download
)
1469 self
._playlist
_level
-= 1
1470 if not self
._playlist
_level
:
1471 self
._playlist
_urls
.clear()
1472 elif result_type
== 'compat_list':
1473 self
.report_warning(
1474 'Extractor %s returned a compat_list result. '
1475 'It needs to be updated.' % ie_result
.get('extractor'))
1478 self
.add_extra_info(r
, {
1479 'extractor': ie_result
['extractor'],
1480 'webpage_url': ie_result
['webpage_url'],
1481 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1482 'extractor_key': ie_result
['extractor_key'],
1485 ie_result
['entries'] = [
1486 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1487 for r
in ie_result
['entries']
1491 raise Exception('Invalid result type: %s' % result_type
)
1493 def _ensure_dir_exists(self
, path
):
1494 return make_dir(path
, self
.report_error
)
1496 def __process_playlist(self
, ie_result
, download
):
1497 # We process each entry in the playlist
1498 playlist
= ie_result
.get('title') or ie_result
.get('id')
1499 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1501 if 'entries' not in ie_result
:
1502 raise EntryNotInPlaylist('There are no entries')
1503 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1504 if incomplete_entries
:
1505 def fill_missing_entries(entries
, indexes
):
1506 ret
= [None] * max(*indexes
)
1507 for i
, entry
in zip(indexes
, entries
):
1510 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1512 playlist_results
= []
1514 playliststart
= self
.params
.get('playliststart', 1)
1515 playlistend
= self
.params
.get('playlistend')
1516 # For backwards compatibility, interpret -1 as whole list
1517 if playlistend
== -1:
1520 playlistitems_str
= self
.params
.get('playlist_items')
1521 playlistitems
= None
1522 if playlistitems_str
is not None:
1523 def iter_playlistitems(format
):
1524 for string_segment
in format
.split(','):
1525 if '-' in string_segment
:
1526 start
, end
= string_segment
.split('-')
1527 for item
in range(int(start
), int(end
) + 1):
1530 yield int(string_segment
)
1531 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1533 ie_entries
= ie_result
['entries']
1535 'Downloading %d videos' if not isinstance(ie_entries
, list)
1536 else 'Collected %d videos; downloading %%d of them' % len(ie_entries
))
1538 if isinstance(ie_entries
, list):
1540 return ie_entries
[i
- 1]
1542 if not isinstance(ie_entries
, (PagedList
, LazyList
)):
1543 ie_entries
= LazyList(ie_entries
)
1546 return YoutubeDL
.__handle
_extraction
_exceptions
(
1547 lambda self
, i
: ie_entries
[i
- 1]
1551 items
= playlistitems
if playlistitems
is not None else itertools
.count(playliststart
)
1555 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1559 entry
= get_entry(i
)
1561 raise EntryNotInPlaylist()
1562 except (IndexError, EntryNotInPlaylist
):
1563 if incomplete_entries
:
1564 raise EntryNotInPlaylist(f
'Entry {i} cannot be found')
1565 elif not playlistitems
:
1567 entries
.append(entry
)
1569 if entry
is not None:
1570 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1571 except (ExistingVideoReached
, RejectedVideoReached
):
1573 ie_result
['entries'] = entries
1575 # Save playlist_index before re-ordering
1577 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1578 for i
, entry
in enumerate(entries
, 1)
1579 if entry
is not None]
1580 n_entries
= len(entries
)
1582 if not playlistitems
and (playliststart
or playlistend
):
1583 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1584 ie_result
['requested_entries'] = playlistitems
1586 if not self
.params
.get('simulate') and self
.params
.get('allow_playlist_files', True):
1588 'playlist': playlist
,
1589 'playlist_id': ie_result
.get('id'),
1590 'playlist_title': ie_result
.get('title'),
1591 'playlist_uploader': ie_result
.get('uploader'),
1592 'playlist_uploader_id': ie_result
.get('uploader_id'),
1593 'playlist_index': 0,
1594 'n_entries': n_entries
,
1596 ie_copy
.update(dict(ie_result
))
1598 if self
._write
_info
_json
('playlist', ie_result
,
1599 self
.prepare_filename(ie_copy
, 'pl_infojson')) is None:
1601 if self
._write
_description
('playlist', ie_result
,
1602 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1604 # TODO: This should be passed to ThumbnailsConvertor if necessary
1605 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1607 if self
.params
.get('playlistreverse', False):
1608 entries
= entries
[::-1]
1609 if self
.params
.get('playlistrandom', False):
1610 random
.shuffle(entries
)
1612 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1614 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1616 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1617 for i
, entry_tuple
in enumerate(entries
, 1):
1618 playlist_index
, entry
= entry_tuple
1619 if 'playlist-index' in self
.params
.get('compat_opts', []):
1620 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1
1621 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1622 # This __x_forwarded_for_ip thing is a bit ugly but requires
1625 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1627 'n_entries': n_entries
,
1628 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1629 'playlist_index': playlist_index
,
1630 'playlist_autonumber': i
,
1631 'playlist': playlist
,
1632 'playlist_id': ie_result
.get('id'),
1633 'playlist_title': ie_result
.get('title'),
1634 'playlist_uploader': ie_result
.get('uploader'),
1635 'playlist_uploader_id': ie_result
.get('uploader_id'),
1636 'extractor': ie_result
['extractor'],
1637 'webpage_url': ie_result
['webpage_url'],
1638 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1639 'extractor_key': ie_result
['extractor_key'],
1642 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1645 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1646 if not entry_result
:
1648 if failures
>= max_failures
:
1650 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1652 # TODO: skip failed (empty) entries?
1653 playlist_results
.append(entry_result
)
1654 ie_result
['entries'] = playlist_results
1655 self
.to_screen('[download] Finished downloading playlist: %s' % playlist
)
1658 @__handle_extraction_exceptions
1659 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1660 return self
.process_ie_result(
1661 entry
, download
=download
, extra_info
=extra_info
)
1663 def _build_format_filter(self
, filter_spec
):
1664 " Returns a function to filter the formats according to the filter_spec "
1674 operator_rex
= re
.compile(r
'''(?x)\s*
1675 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1676 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1677 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1678 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1679 m
= operator_rex
.fullmatch(filter_spec
)
1682 comparison_value
= int(m
.group('value'))
1684 comparison_value
= parse_filesize(m
.group('value'))
1685 if comparison_value
is None:
1686 comparison_value
= parse_filesize(m
.group('value') + 'B')
1687 if comparison_value
is None:
1689 'Invalid value %r in format specification %r' % (
1690 m
.group('value'), filter_spec
))
1691 op
= OPERATORS
[m
.group('op')]
1696 '^=': lambda attr
, value
: attr
.startswith(value
),
1697 '$=': lambda attr
, value
: attr
.endswith(value
),
1698 '*=': lambda attr
, value
: value
in attr
,
1700 str_operator_rex
= re
.compile(r
'''(?x)\s*
1701 (?P<key>[a-zA-Z0-9._-]+)\s*
1702 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1703 (?P<value>[a-zA-Z0-9._-]+)\s*
1704 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1705 m
= str_operator_rex
.fullmatch(filter_spec
)
1707 comparison_value
= m
.group('value')
1708 str_op
= STR_OPERATORS
[m
.group('op')]
1709 if m
.group('negation'):
1710 op
= lambda attr
, value
: not str_op(attr
, value
)
1715 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1718 actual_value
= f
.get(m
.group('key'))
1719 if actual_value
is None:
1720 return m
.group('none_inclusive')
1721 return op(actual_value
, comparison_value
)
1724 def _check_formats(self
, formats
):
1726 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1727 temp_file
= tempfile
.NamedTemporaryFile(
1728 suffix
='.tmp', delete
=False,
1729 dir=self
.get_output_path('temp') or None)
1732 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1733 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1736 if os
.path
.exists(temp_file
.name
):
1738 os
.remove(temp_file
.name
)
1740 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1744 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1746 def _default_format_spec(self
, info_dict
, download
=True):
1749 merger
= FFmpegMergerPP(self
)
1750 return merger
.available
and merger
.can_merge()
1753 not self
.params
.get('simulate')
1757 or info_dict
.get('is_live', False)
1758 or self
.outtmpl_dict
['default'] == '-'))
1761 or self
.params
.get('allow_multiple_audio_streams', False)
1762 or 'format-spec' in self
.params
.get('compat_opts', []))
1765 'best/bestvideo+bestaudio' if prefer_best
1766 else 'bestvideo*+bestaudio/best' if not compat
1767 else 'bestvideo+bestaudio/best')
1769 def build_format_selector(self
, format_spec
):
1770 def syntax_error(note
, start
):
1772 'Invalid format specification: '
1773 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1774 return SyntaxError(message
)
1776 PICKFIRST
= 'PICKFIRST'
1780 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1782 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1783 'video': self
.params
.get('allow_multiple_video_streams', False)}
1785 check_formats
= self
.params
.get('check_formats') == 'selected'
1787 def _parse_filter(tokens
):
1789 for type, string
, start
, _
, _
in tokens
:
1790 if type == tokenize
.OP
and string
== ']':
1791 return ''.join(filter_parts
)
1793 filter_parts
.append(string
)
1795 def _remove_unused_ops(tokens
):
1796 # Remove operators that we don't use and join them with the surrounding strings
1797 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1798 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1799 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1800 for type, string
, start
, end
, line
in tokens
:
1801 if type == tokenize
.OP
and string
== '[':
1803 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1805 yield type, string
, start
, end
, line
1806 # everything inside brackets will be handled by _parse_filter
1807 for type, string
, start
, end
, line
in tokens
:
1808 yield type, string
, start
, end
, line
1809 if type == tokenize
.OP
and string
== ']':
1811 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1813 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1815 yield type, string
, start
, end
, line
1816 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1818 last_string
= string
1822 last_string
+= string
1824 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1826 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1828 current_selector
= None
1829 for type, string
, start
, _
, _
in tokens
:
1830 # ENCODING is only defined in python 3.x
1831 if type == getattr(tokenize
, 'ENCODING', None):
1833 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1834 current_selector
= FormatSelector(SINGLE
, string
, [])
1835 elif type == tokenize
.OP
:
1837 if not inside_group
:
1838 # ')' will be handled by the parentheses group
1839 tokens
.restore_last_token()
1841 elif inside_merge
and string
in ['/', ',']:
1842 tokens
.restore_last_token()
1844 elif inside_choice
and string
== ',':
1845 tokens
.restore_last_token()
1848 if not current_selector
:
1849 raise syntax_error('"," must follow a format selector', start
)
1850 selectors
.append(current_selector
)
1851 current_selector
= None
1853 if not current_selector
:
1854 raise syntax_error('"/" must follow a format selector', start
)
1855 first_choice
= current_selector
1856 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1857 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1859 if not current_selector
:
1860 current_selector
= FormatSelector(SINGLE
, 'best', [])
1861 format_filter
= _parse_filter(tokens
)
1862 current_selector
.filters
.append(format_filter
)
1864 if current_selector
:
1865 raise syntax_error('Unexpected "("', start
)
1866 group
= _parse_format_selection(tokens
, inside_group
=True)
1867 current_selector
= FormatSelector(GROUP
, group
, [])
1869 if not current_selector
:
1870 raise syntax_error('Unexpected "+"', start
)
1871 selector_1
= current_selector
1872 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1874 raise syntax_error('Expected a selector', start
)
1875 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
1877 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
1878 elif type == tokenize
.ENDMARKER
:
1880 if current_selector
:
1881 selectors
.append(current_selector
)
1884 def _merge(formats_pair
):
1885 format_1
, format_2
= formats_pair
1888 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
1889 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
1891 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
1892 get_no_more
= {'video': False, 'audio': False}
1893 for (i
, fmt_info
) in enumerate(formats_info
):
1894 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
1897 for aud_vid
in ['audio', 'video']:
1898 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
1899 if get_no_more
[aud_vid
]:
1902 get_no_more
[aud_vid
] = True
1904 if len(formats_info
) == 1:
1905 return formats_info
[0]
1907 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
1908 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
1910 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
1911 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
1913 output_ext
= self
.params
.get('merge_output_format')
1916 output_ext
= the_only_video
['ext']
1917 elif the_only_audio
and not video_fmts
:
1918 output_ext
= the_only_audio
['ext']
1922 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
1925 'requested_formats': formats_info
,
1926 'format': '+'.join(filtered('format')),
1927 'format_id': '+'.join(filtered('format_id')),
1929 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
1930 'language': '+'.join(orderedSet(filtered('language'))),
1931 'format_note': '+'.join(orderedSet(filtered('format_note'))),
1932 'filesize_approx': sum(filtered('filesize', 'filesize_approx')),
1933 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
1938 'width': the_only_video
.get('width'),
1939 'height': the_only_video
.get('height'),
1940 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
1941 'fps': the_only_video
.get('fps'),
1942 'dynamic_range': the_only_video
.get('dynamic_range'),
1943 'vcodec': the_only_video
.get('vcodec'),
1944 'vbr': the_only_video
.get('vbr'),
1945 'stretched_ratio': the_only_video
.get('stretched_ratio'),
1950 'acodec': the_only_audio
.get('acodec'),
1951 'abr': the_only_audio
.get('abr'),
1952 'asr': the_only_audio
.get('asr'),
1957 def _check_formats(formats
):
1958 if not check_formats
:
1961 yield from self
._check
_formats
(formats
)
1963 def _build_selector_function(selector
):
1964 if isinstance(selector
, list): # ,
1965 fs
= [_build_selector_function(s
) for s
in selector
]
1967 def selector_function(ctx
):
1970 return selector_function
1972 elif selector
.type == GROUP
: # ()
1973 selector_function
= _build_selector_function(selector
.selector
)
1975 elif selector
.type == PICKFIRST
: # /
1976 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
1978 def selector_function(ctx
):
1980 picked_formats
= list(f(ctx
))
1982 return picked_formats
1985 elif selector
.type == MERGE
: # +
1986 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
1988 def selector_function(ctx
):
1989 for pair
in itertools
.product(
1990 selector_1(copy
.deepcopy(ctx
)), selector_2(copy
.deepcopy(ctx
))):
1993 elif selector
.type == SINGLE
: # atom
1994 format_spec
= selector
.selector
or 'best'
1996 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
1997 if format_spec
== 'all':
1998 def selector_function(ctx
):
1999 yield from _check_formats(ctx
['formats'])
2000 elif format_spec
== 'mergeall':
2001 def selector_function(ctx
):
2002 formats
= list(_check_formats(ctx
['formats']))
2005 merged_format
= formats
[-1]
2006 for f
in formats
[-2::-1]:
2007 merged_format
= _merge((merged_format
, f
))
2011 format_fallback
, format_reverse
, format_idx
= False, True, 1
2013 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2015 if mobj
is not None:
2016 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2017 format_reverse
= mobj
.group('bw')[0] == 'b'
2018 format_type
= (mobj
.group('type') or [None])[0]
2019 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2020 format_modified
= mobj
.group('mod') is not None
2022 format_fallback
= not format_type
and not format_modified
# for b, w
2024 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2025 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2026 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2027 if format_type
# bv, ba, wv, wa
2028 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2029 if not format_modified
# b, w
2030 else lambda f
: True) # b*, w*
2031 filter_f
= lambda f
: _filter_f(f
) and (
2032 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2034 if format_spec
in self
._format
_selection
_exts
['audio']:
2035 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2036 elif format_spec
in self
._format
_selection
_exts
['video']:
2037 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2038 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2039 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2041 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2043 def selector_function(ctx
):
2044 formats
= list(ctx
['formats'])
2045 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2046 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
2047 # for extractors with incomplete formats (audio only (soundcloud)
2048 # or video only (imgur)) best/worst will fallback to
2049 # best/worst {video,audio}-only format
2051 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2053 yield matches
[format_idx
- 1]
2057 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2059 def final_selector(ctx
):
2060 ctx_copy
= copy
.deepcopy(ctx
)
2061 for _filter
in filters
:
2062 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2063 return selector_function(ctx_copy
)
2064 return final_selector
2066 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
2068 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
2069 except tokenize
.TokenError
:
2070 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2072 class TokenIterator(object):
2073 def __init__(self
, tokens
):
2074 self
.tokens
= tokens
2081 if self
.counter
>= len(self
.tokens
):
2082 raise StopIteration()
2083 value
= self
.tokens
[self
.counter
]
2089 def restore_last_token(self
):
2092 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2093 return _build_selector_function(parsed_selector
)
2095 def _calc_headers(self
, info_dict
):
2096 res
= std_headers
.copy()
2098 add_headers
= info_dict
.get('http_headers')
2100 res
.update(add_headers
)
2102 cookies
= self
._calc
_cookies
(info_dict
)
2104 res
['Cookie'] = cookies
2106 if 'X-Forwarded-For' not in res
:
2107 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2108 if x_forwarded_for_ip
:
2109 res
['X-Forwarded-For'] = x_forwarded_for_ip
2113 def _calc_cookies(self
, info_dict
):
2114 pr
= sanitized_Request(info_dict
['url'])
2115 self
.cookiejar
.add_cookie_header(pr
)
2116 return pr
.get_header('Cookie')
2118 def _sort_thumbnails(self
, thumbnails
):
2119 thumbnails
.sort(key
=lambda t
: (
2120 t
.get('preference') if t
.get('preference') is not None else -1,
2121 t
.get('width') if t
.get('width') is not None else -1,
2122 t
.get('height') if t
.get('height') is not None else -1,
2123 t
.get('id') if t
.get('id') is not None else '',
2126 def _sanitize_thumbnails(self
, info_dict
):
2127 thumbnails
= info_dict
.get('thumbnails')
2128 if thumbnails
is None:
2129 thumbnail
= info_dict
.get('thumbnail')
2131 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2135 def check_thumbnails(thumbnails
):
2136 for t
in thumbnails
:
2137 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2139 self
.urlopen(HEADRequest(t
['url']))
2140 except network_exceptions
as err
:
2141 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2145 self
._sort
_thumbnails
(thumbnails
)
2146 for i
, t
in enumerate(thumbnails
):
2147 if t
.get('id') is None:
2149 if t
.get('width') and t
.get('height'):
2150 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2151 t
['url'] = sanitize_url(t
['url'])
2153 if self
.params
.get('check_formats') is True:
2154 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1])).reverse()
2156 info_dict
['thumbnails'] = thumbnails
2158 def process_video_result(self
, info_dict
, download
=True):
2159 assert info_dict
.get('_type', 'video') == 'video'
2161 if 'id' not in info_dict
:
2162 raise ExtractorError('Missing "id" field in extractor result')
2163 if 'title' not in info_dict
:
2164 raise ExtractorError('Missing "title" field in extractor result',
2165 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2167 def report_force_conversion(field
, field_not
, conversion
):
2168 self
.report_warning(
2169 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2170 % (field
, field_not
, conversion
))
2172 def sanitize_string_field(info
, string_field
):
2173 field
= info
.get(string_field
)
2174 if field
is None or isinstance(field
, compat_str
):
2176 report_force_conversion(string_field
, 'a string', 'string')
2177 info
[string_field
] = compat_str(field
)
2179 def sanitize_numeric_fields(info
):
2180 for numeric_field
in self
._NUMERIC
_FIELDS
:
2181 field
= info
.get(numeric_field
)
2182 if field
is None or isinstance(field
, compat_numeric_types
):
2184 report_force_conversion(numeric_field
, 'numeric', 'int')
2185 info
[numeric_field
] = int_or_none(field
)
2187 sanitize_string_field(info_dict
, 'id')
2188 sanitize_numeric_fields(info_dict
)
2190 if 'playlist' not in info_dict
:
2191 # It isn't part of a playlist
2192 info_dict
['playlist'] = None
2193 info_dict
['playlist_index'] = None
2195 self
._sanitize
_thumbnails
(info_dict
)
2197 thumbnail
= info_dict
.get('thumbnail')
2198 thumbnails
= info_dict
.get('thumbnails')
2200 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2202 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2204 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2205 info_dict
['display_id'] = info_dict
['id']
2207 if info_dict
.get('duration') is not None:
2208 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2210 for ts_key
, date_key
in (
2211 ('timestamp', 'upload_date'),
2212 ('release_timestamp', 'release_date'),
2214 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2215 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2216 # see http://bugs.python.org/issue1646728)
2218 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2219 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2220 except (ValueError, OverflowError, OSError):
2223 live_keys
= ('is_live', 'was_live')
2224 live_status
= info_dict
.get('live_status')
2225 if live_status
is None:
2226 for key
in live_keys
:
2227 if info_dict
.get(key
) is False:
2229 if info_dict
.get(key
):
2232 if all(info_dict
.get(key
) is False for key
in live_keys
):
2233 live_status
= 'not_live'
2235 info_dict
['live_status'] = live_status
2236 for key
in live_keys
:
2237 if info_dict
.get(key
) is None:
2238 info_dict
[key
] = (live_status
== key
)
2240 # Auto generate title fields corresponding to the *_number fields when missing
2241 # in order to always have clean titles. This is very common for TV series.
2242 for field
in ('chapter', 'season', 'episode'):
2243 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2244 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2246 for cc_kind
in ('subtitles', 'automatic_captions'):
2247 cc
= info_dict
.get(cc_kind
)
2249 for _
, subtitle
in cc
.items():
2250 for subtitle_format
in subtitle
:
2251 if subtitle_format
.get('url'):
2252 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2253 if subtitle_format
.get('ext') is None:
2254 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2256 automatic_captions
= info_dict
.get('automatic_captions')
2257 subtitles
= info_dict
.get('subtitles')
2259 info_dict
['requested_subtitles'] = self
.process_subtitles(
2260 info_dict
['id'], subtitles
, automatic_captions
)
2262 if info_dict
.get('formats') is None:
2263 # There's only one format available
2264 formats
= [info_dict
]
2266 formats
= info_dict
['formats']
2268 info_dict
['__has_drm'] = any(f
.get('has_drm') for f
in formats
)
2269 if not self
.params
.get('allow_unplayable_formats'):
2270 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2273 self
.raise_no_formats(info_dict
)
2275 def is_wellformed(f
):
2278 self
.report_warning(
2279 '"url" field is missing or empty - skipping format, '
2280 'there is an error in extractor')
2282 if isinstance(url
, bytes):
2283 sanitize_string_field(f
, 'url')
2286 # Filter out malformed formats for better extraction robustness
2287 formats
= list(filter(is_wellformed
, formats
))
2291 # We check that all the formats have the format and format_id fields
2292 for i
, format
in enumerate(formats
):
2293 sanitize_string_field(format
, 'format_id')
2294 sanitize_numeric_fields(format
)
2295 format
['url'] = sanitize_url(format
['url'])
2296 if not format
.get('format_id'):
2297 format
['format_id'] = compat_str(i
)
2299 # Sanitize format_id from characters used in format selector expression
2300 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2301 format_id
= format
['format_id']
2302 if format_id
not in formats_dict
:
2303 formats_dict
[format_id
] = []
2304 formats_dict
[format_id
].append(format
)
2306 # Make sure all formats have unique format_id
2307 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2308 for format_id
, ambiguous_formats
in formats_dict
.items():
2309 ambigious_id
= len(ambiguous_formats
) > 1
2310 for i
, format
in enumerate(ambiguous_formats
):
2312 format
['format_id'] = '%s-%d' % (format_id
, i
)
2313 if format
.get('ext') is None:
2314 format
['ext'] = determine_ext(format
['url']).lower()
2315 # Ensure there is no conflict between id and ext in format selection
2316 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2317 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2318 format
['format_id'] = 'f%s' % format
['format_id']
2320 for i
, format
in enumerate(formats
):
2321 if format
.get('format') is None:
2322 format
['format'] = '{id} - {res}{note}'.format(
2323 id=format
['format_id'],
2324 res
=self
.format_resolution(format
),
2325 note
=format_field(format
, 'format_note', ' (%s)'),
2327 if format
.get('protocol') is None:
2328 format
['protocol'] = determine_protocol(format
)
2329 if format
.get('resolution') is None:
2330 format
['resolution'] = self
.format_resolution(format
, default
=None)
2331 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2332 format
['dynamic_range'] = 'SDR'
2333 if (info_dict
.get('duration') and format
.get('tbr')
2334 and not format
.get('filesize') and not format
.get('filesize_approx')):
2335 format
['filesize_approx'] = info_dict
['duration'] * format
['tbr'] * (1024 / 8)
2337 # Add HTTP headers, so that external programs can use them from the
2339 full_format_info
= info_dict
.copy()
2340 full_format_info
.update(format
)
2341 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2342 # Remove private housekeeping stuff
2343 if '__x_forwarded_for_ip' in info_dict
:
2344 del info_dict
['__x_forwarded_for_ip']
2346 # TODO Central sorting goes here
2348 if self
.params
.get('check_formats') is True:
2349 formats
= LazyList(self
._check
_formats
(formats
[::-1])).reverse()
2351 if not formats
or formats
[0] is not info_dict
:
2352 # only set the 'formats' fields if the original info_dict list them
2353 # otherwise we end up with a circular reference, the first (and unique)
2354 # element in the 'formats' field in info_dict is info_dict itself,
2355 # which can't be exported to json
2356 info_dict
['formats'] = formats
2358 info_dict
, _
= self
.pre_process(info_dict
)
2360 if self
.params
.get('list_thumbnails'):
2361 self
.list_thumbnails(info_dict
)
2362 if self
.params
.get('listformats'):
2363 if not info_dict
.get('formats') and not info_dict
.get('url'):
2364 self
.to_screen('%s has no formats' % info_dict
['id'])
2366 self
.list_formats(info_dict
)
2367 if self
.params
.get('listsubtitles'):
2368 if 'automatic_captions' in info_dict
:
2369 self
.list_subtitles(
2370 info_dict
['id'], automatic_captions
, 'automatic captions')
2371 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2372 list_only
= self
.params
.get('simulate') is None and (
2373 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2375 # Without this printing, -F --print-json will not work
2376 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2379 format_selector
= self
.format_selector
2380 if format_selector
is None:
2381 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2382 self
.write_debug('Default format spec: %s' % req_format
)
2383 format_selector
= self
.build_format_selector(req_format
)
2385 # While in format selection we may need to have an access to the original
2386 # format set in order to calculate some metrics or do some processing.
2387 # For now we need to be able to guess whether original formats provided
2388 # by extractor are incomplete or not (i.e. whether extractor provides only
2389 # video-only or audio-only formats) for proper formats selection for
2390 # extractors with such incomplete formats (see
2391 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2392 # Since formats may be filtered during format selection and may not match
2393 # the original formats the results may be incorrect. Thus original formats
2394 # or pre-calculated metrics should be passed to format selection routines
2396 # We will pass a context object containing all necessary additional data
2397 # instead of just formats.
2398 # This fixes incorrect format selection issue (see
2399 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2400 incomplete_formats
= (
2401 # All formats are video-only or
2402 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2403 # all formats are audio-only
2404 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2408 'incomplete_formats': incomplete_formats
,
2411 formats_to_download
= list(format_selector(ctx
))
2412 if not formats_to_download
:
2413 if not self
.params
.get('ignore_no_formats_error'):
2414 raise ExtractorError('Requested format is not available', expected
=True,
2415 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2417 self
.report_warning('Requested format is not available')
2418 # Process what we can, even without any available formats.
2419 self
.process_info(dict(info_dict
))
2422 '[info] %s: Downloading %d format(s): %s' % (
2423 info_dict
['id'], len(formats_to_download
),
2424 ", ".join([f
['format_id'] for f
in formats_to_download
])))
2425 for fmt
in formats_to_download
:
2426 new_info
= dict(info_dict
)
2427 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2428 new_info
['__original_infodict'] = info_dict
2429 new_info
.update(fmt
)
2430 self
.process_info(new_info
)
2431 # We update the info dict with the selected best quality format (backwards compatibility)
2432 if formats_to_download
:
2433 info_dict
.update(formats_to_download
[-1])
2436 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2437 """Select the requested subtitles and their format"""
2439 if normal_subtitles
and self
.params
.get('writesubtitles'):
2440 available_subs
.update(normal_subtitles
)
2441 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2442 for lang
, cap_info
in automatic_captions
.items():
2443 if lang
not in available_subs
:
2444 available_subs
[lang
] = cap_info
2446 if (not self
.params
.get('writesubtitles') and not
2447 self
.params
.get('writeautomaticsub') or not
2451 all_sub_langs
= available_subs
.keys()
2452 if self
.params
.get('allsubtitles', False):
2453 requested_langs
= all_sub_langs
2454 elif self
.params
.get('subtitleslangs', False):
2455 # A list is used so that the order of languages will be the same as
2456 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2457 requested_langs
= []
2458 for lang_re
in self
.params
.get('subtitleslangs'):
2459 if lang_re
== 'all':
2460 requested_langs
.extend(all_sub_langs
)
2462 discard
= lang_re
[0] == '-'
2464 lang_re
= lang_re
[1:]
2465 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2467 for lang
in current_langs
:
2468 while lang
in requested_langs
:
2469 requested_langs
.remove(lang
)
2471 requested_langs
.extend(current_langs
)
2472 requested_langs
= orderedSet(requested_langs
)
2473 elif 'en' in available_subs
:
2474 requested_langs
= ['en']
2476 requested_langs
= [list(all_sub_langs
)[0]]
2478 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2480 formats_query
= self
.params
.get('subtitlesformat', 'best')
2481 formats_preference
= formats_query
.split('/') if formats_query
else []
2483 for lang
in requested_langs
:
2484 formats
= available_subs
.get(lang
)
2486 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2488 for ext
in formats_preference
:
2492 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2498 self
.report_warning(
2499 'No subtitle format found matching "%s" for language %s, '
2500 'using %s' % (formats_query
, lang
, f
['ext']))
2504 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2505 def print_mandatory(field
, actual_field
=None):
2506 if actual_field
is None:
2507 actual_field
= field
2508 if (self
.params
.get('force%s' % field
, False)
2509 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2510 self
.to_stdout(info_dict
[actual_field
])
2512 def print_optional(field
):
2513 if (self
.params
.get('force%s' % field
, False)
2514 and info_dict
.get(field
) is not None):
2515 self
.to_stdout(info_dict
[field
])
2517 info_dict
= info_dict
.copy()
2518 if filename
is not None:
2519 info_dict
['filename'] = filename
2520 if info_dict
.get('requested_formats') is not None:
2521 # For RTMP URLs, also include the playpath
2522 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2523 elif 'url' in info_dict
:
2524 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2526 if self
.params
.get('forceprint') or self
.params
.get('forcejson'):
2527 self
.post_extract(info_dict
)
2528 for tmpl
in self
.params
.get('forceprint', []):
2529 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2530 if mobj
and mobj
.group(1):
2531 tmpl
= f
'{tmpl[:-1]} = %({tmpl[:-1]})s'
2533 tmpl
= '%({})s'.format(tmpl
)
2534 self
.to_stdout(self
.evaluate_outtmpl(tmpl
, info_dict
))
2536 print_mandatory('title')
2537 print_mandatory('id')
2538 print_mandatory('url', 'urls')
2539 print_optional('thumbnail')
2540 print_optional('description')
2541 print_optional('filename')
2542 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2543 self
.to_stdout(formatSeconds(info_dict
['duration']))
2544 print_mandatory('format')
2546 if self
.params
.get('forcejson'):
2547 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2549 def dl(self
, name
, info
, subtitle
=False, test
=False):
2550 if not info
.get('url'):
2551 self
.raise_no_formats(info
, True)
2554 verbose
= self
.params
.get('verbose')
2557 'quiet': self
.params
.get('quiet') or not verbose
,
2559 'noprogress': not verbose
,
2561 'skip_unavailable_fragments': False,
2562 'keep_fragments': False,
2564 '_no_ytdl_file': True,
2567 params
= self
.params
2568 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2570 for ph
in self
._progress
_hooks
:
2571 fd
.add_progress_hook(ph
)
2572 urls
= '", "'.join([f
['url'] for f
in info
.get('requested_formats', [])] or [info
['url']])
2573 self
.write_debug('Invoking downloader on "%s"' % urls
)
2575 new_info
= copy
.deepcopy(self
._copy
_infodict
(info
))
2576 if new_info
.get('http_headers') is None:
2577 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2578 return fd
.download(name
, new_info
, subtitle
)
2580 def process_info(self
, info_dict
):
2581 """Process a single resolved IE result."""
2583 assert info_dict
.get('_type', 'video') == 'video'
2585 max_downloads
= self
.params
.get('max_downloads')
2586 if max_downloads
is not None:
2587 if self
._num
_downloads
>= int(max_downloads
):
2588 raise MaxDownloadsReached()
2590 # TODO: backward compatibility, to be removed
2591 info_dict
['fulltitle'] = info_dict
['title']
2593 if 'format' not in info_dict
and 'ext' in info_dict
:
2594 info_dict
['format'] = info_dict
['ext']
2596 if self
._match
_entry
(info_dict
) is not None:
2599 self
.post_extract(info_dict
)
2600 self
._num
_downloads
+= 1
2602 # info_dict['_filename'] needs to be set for backward compatibility
2603 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2604 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2608 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2610 if self
.params
.get('simulate'):
2611 if self
.params
.get('force_write_download_archive', False):
2612 self
.record_download_archive(info_dict
)
2613 # Do nothing else if in simulate mode
2616 if full_filename
is None:
2618 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2620 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2623 if self
._write
_description
('video', info_dict
,
2624 self
.prepare_filename(info_dict
, 'description')) is None:
2627 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2628 if sub_files
is None:
2630 files_to_move
.update(dict(sub_files
))
2632 thumb_files
= self
._write
_thumbnails
(
2633 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2634 if thumb_files
is None:
2636 files_to_move
.update(dict(thumb_files
))
2638 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2639 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2640 if _infojson_written
:
2641 info_dict
['__infojson_filename'] = infofn
2642 elif _infojson_written
is None:
2645 # Note: Annotations are deprecated
2647 if self
.params
.get('writeannotations', False):
2648 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2650 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2652 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2653 self
.to_screen('[info] Video annotations are already present')
2654 elif not info_dict
.get('annotations'):
2655 self
.report_warning('There are no annotations to write.')
2658 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2659 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2660 annofile
.write(info_dict
['annotations'])
2661 except (KeyError, TypeError):
2662 self
.report_warning('There are no annotations to write.')
2663 except (OSError, IOError):
2664 self
.report_error('Cannot write annotations file: ' + annofn
)
2667 # Write internet shortcut files
2668 def _write_link_file(link_type
):
2669 if 'webpage_url' not in info_dict
:
2670 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2672 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2673 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2674 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2677 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2678 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2679 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2680 template_vars
= {'url': iri_to_uri(info_dict['webpage_url'])}
2681 if link_type
== 'desktop':
2682 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2683 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
2684 except (OSError, IOError):
2685 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
2690 'url': self
.params
.get('writeurllink'),
2691 'webloc': self
.params
.get('writewebloclink'),
2692 'desktop': self
.params
.get('writedesktoplink'),
2694 if self
.params
.get('writelink'):
2695 link_type
= ('webloc' if sys
.platform
== 'darwin'
2696 else 'desktop' if sys
.platform
.startswith('linux')
2698 write_links
[link_type
] = True
2700 if any(should_write
and not _write_link_file(link_type
)
2701 for link_type
, should_write
in write_links
.items()):
2705 info_dict
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2706 except PostProcessingError
as err
:
2707 self
.report_error('Preprocessing: %s' % str(err
))
2710 must_record_download_archive
= False
2711 if self
.params
.get('skip_download', False):
2712 info_dict
['filepath'] = temp_filename
2713 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2714 info_dict
['__files_to_move'] = files_to_move
2715 info_dict
= self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
)
2718 info_dict
.setdefault('__postprocessors', [])
2721 def existing_file(*filepaths
):
2722 ext
= info_dict
.get('ext')
2723 final_ext
= self
.params
.get('final_ext', ext
)
2725 for file in orderedSet(filepaths
):
2726 if final_ext
!= ext
:
2727 converted
= replace_extension(file, final_ext
, ext
)
2728 if os
.path
.exists(encodeFilename(converted
)):
2729 existing_files
.append(converted
)
2730 if os
.path
.exists(encodeFilename(file)):
2731 existing_files
.append(file)
2733 if not existing_files
or self
.params
.get('overwrites', False):
2734 for file in orderedSet(existing_files
):
2735 self
.report_file_delete(file)
2736 os
.remove(encodeFilename(file))
2739 info_dict
['ext'] = os
.path
.splitext(existing_files
[0])[1][1:]
2740 return existing_files
[0]
2743 if info_dict
.get('requested_formats') is not None:
2745 def compatible_formats(formats
):
2746 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2747 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2748 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2749 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2753 exts
= set(format
.get('ext') for format
in formats
)
2755 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2758 for ext_sets
in COMPATIBLE_EXTS
:
2759 if ext_sets
.issuperset(exts
):
2761 # TODO: Check acodec/vcodec
2764 requested_formats
= info_dict
['requested_formats']
2765 old_ext
= info_dict
['ext']
2766 if self
.params
.get('merge_output_format') is None:
2767 if not compatible_formats(requested_formats
):
2768 info_dict
['ext'] = 'mkv'
2769 self
.report_warning(
2770 'Requested formats are incompatible for merge and will be merged into mkv')
2771 if (info_dict
['ext'] == 'webm'
2772 and info_dict
.get('thumbnails')
2773 # check with type instead of pp_key, __name__, or isinstance
2774 # since we dont want any custom PPs to trigger this
2775 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])):
2776 info_dict
['ext'] = 'mkv'
2777 self
.report_warning(
2778 'webm doesn\'t support embedding a thumbnail, mkv will be used')
2779 new_ext
= info_dict
['ext']
2781 def correct_ext(filename
, ext
=new_ext
):
2784 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2786 os
.path
.splitext(filename
)[0]
2787 if filename_real_ext
in (old_ext
, new_ext
)
2789 return '%s.%s' % (filename_wo_ext
, ext
)
2791 # Ensure filename always has a correct extension for successful merge
2792 full_filename
= correct_ext(full_filename
)
2793 temp_filename
= correct_ext(temp_filename
)
2794 dl_filename
= existing_file(full_filename
, temp_filename
)
2795 info_dict
['__real_download'] = False
2797 if dl_filename
is not None:
2798 self
.report_file_already_downloaded(dl_filename
)
2799 elif get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-'):
2800 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
2801 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2802 info_dict
['__real_download'] = real_download
2805 merger
= FFmpegMergerPP(self
)
2806 if self
.params
.get('allow_unplayable_formats'):
2807 self
.report_warning(
2808 'You have requested merging of multiple formats '
2809 'while also allowing unplayable formats to be downloaded. '
2810 'The formats won\'t be merged to prevent data corruption.')
2811 elif not merger
.available
:
2812 self
.report_warning(
2813 'You have requested merging of multiple formats but ffmpeg is not installed. '
2814 'The formats won\'t be merged.')
2816 if temp_filename
== '-':
2817 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
)
2818 else 'but the formats are incompatible for simultaneous download' if merger
.available
2819 else 'but ffmpeg is not installed')
2820 self
.report_warning(
2821 f
'You have requested downloading multiple formats to stdout {reason}. '
2822 'The formats will be streamed one after the other')
2823 fname
= temp_filename
2824 for f
in requested_formats
:
2825 new_info
= dict(info_dict
)
2826 del new_info
['requested_formats']
2828 if temp_filename
!= '-':
2829 fname
= prepend_extension(
2830 correct_ext(temp_filename
, new_info
['ext']),
2831 'f%s' % f
['format_id'], new_info
['ext'])
2832 if not self
._ensure
_dir
_exists
(fname
):
2834 f
['filepath'] = fname
2835 downloaded
.append(fname
)
2836 partial_success
, real_download
= self
.dl(fname
, new_info
)
2837 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
2838 success
= success
and partial_success
2839 if merger
.available
and not self
.params
.get('allow_unplayable_formats'):
2840 info_dict
['__postprocessors'].append(merger
)
2841 info_dict
['__files_to_merge'] = downloaded
2842 # Even if there were no downloads, it is being merged only now
2843 info_dict
['__real_download'] = True
2845 for file in downloaded
:
2846 files_to_move
[file] = None
2848 # Just a single file
2849 dl_filename
= existing_file(full_filename
, temp_filename
)
2850 if dl_filename
is None or dl_filename
== temp_filename
:
2851 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
2852 # So we should try to resume the download
2853 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2854 info_dict
['__real_download'] = real_download
2856 self
.report_file_already_downloaded(dl_filename
)
2858 dl_filename
= dl_filename
or temp_filename
2859 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2861 except network_exceptions
as err
:
2862 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
2864 except (OSError, IOError) as err
:
2865 raise UnavailableVideoError(err
)
2866 except (ContentTooShortError
, ) as err
:
2867 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
2870 if success
and full_filename
!= '-':
2874 fixup_policy
= self
.params
.get('fixup')
2875 vid
= info_dict
['id']
2877 if fixup_policy
in ('ignore', 'never'):
2879 elif fixup_policy
== 'warn':
2881 elif fixup_policy
!= 'force':
2882 assert fixup_policy
in ('detect_or_warn', None)
2883 if not info_dict
.get('__real_download'):
2886 def ffmpeg_fixup(cndn
, msg
, cls
):
2890 self
.report_warning(f
'{vid}: {msg}')
2894 info_dict
['__postprocessors'].append(pp
)
2896 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
2898 stretched_ratio
= info_dict
.get('stretched_ratio')
2900 stretched_ratio
not in (1, None),
2901 f
'Non-uniform pixel ratio {stretched_ratio}',
2902 FFmpegFixupStretchedPP
)
2905 (info_dict
.get('requested_formats') is None
2906 and info_dict
.get('container') == 'm4a_dash'
2907 and info_dict
.get('ext') == 'm4a'),
2908 'writing DASH m4a. Only some players support this container',
2911 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
2912 downloader
= downloader
.__name
__ if downloader
else None
2913 ffmpeg_fixup(info_dict
.get('requested_formats') is None and downloader
== 'HlsFD',
2914 'malformed AAC bitstream detected', FFmpegFixupM3u8PP
)
2915 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP
)
2916 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP
)
2920 info_dict
= self
.post_process(dl_filename
, info_dict
, files_to_move
)
2921 except PostProcessingError
as err
:
2922 self
.report_error('Postprocessing: %s' % str(err
))
2925 for ph
in self
._post
_hooks
:
2926 ph(info_dict
['filepath'])
2927 except Exception as err
:
2928 self
.report_error('post hooks: %s' % str(err
))
2930 must_record_download_archive
= True
2932 if must_record_download_archive
or self
.params
.get('force_write_download_archive', False):
2933 self
.record_download_archive(info_dict
)
2934 max_downloads
= self
.params
.get('max_downloads')
2935 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
2936 raise MaxDownloadsReached()
2938 def __download_wrapper(self
, func
):
2939 @functools.wraps(func
)
2940 def wrapper(*args
, **kwargs
):
2942 res
= func(*args
, **kwargs
)
2943 except UnavailableVideoError
as e
:
2944 self
.report_error(e
)
2945 except DownloadCancelled
as e
:
2946 self
.to_screen(f
'[info] {e}')
2949 if self
.params
.get('dump_single_json', False):
2950 self
.post_extract(res
)
2951 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
2954 def download(self
, url_list
):
2955 """Download a given list of URLs."""
2956 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
2957 outtmpl
= self
.outtmpl_dict
['default']
2958 if (len(url_list
) > 1
2960 and '%' not in outtmpl
2961 and self
.params
.get('max_downloads') != 1):
2962 raise SameFileError(outtmpl
)
2964 for url
in url_list
:
2965 self
.__download
_wrapper
(self
.extract_info
)(
2966 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
2968 return self
._download
_retcode
2970 def download_with_info_file(self
, info_filename
):
2971 with contextlib
.closing(fileinput
.FileInput(
2972 [info_filename
], mode
='r',
2973 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
2974 # FileInput doesn't have a read method, we can't call json.load
2975 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
2977 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
2978 except (DownloadError
, EntryNotInPlaylist
, ThrottledDownload
) as e
:
2979 self
.to_stderr('\r')
2980 webpage_url
= info
.get('webpage_url')
2981 if webpage_url
is not None:
2982 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
2983 return self
.download([webpage_url
])
2986 return self
._download
_retcode
2989 def sanitize_info(info_dict
, remove_private_keys
=False):
2990 ''' Sanitize the infodict for converting to json '''
2991 if info_dict
is None:
2993 info_dict
.setdefault('epoch', int(time
.time()))
2994 remove_keys
= {'__original_infodict'}
# Always remove this since this may contain a copy of the entire dict
2995 keep_keys
= ['_type'] # Always keep this to facilitate load-info-json
2996 if remove_private_keys
:
2998 'requested_formats', 'requested_subtitles', 'requested_entries',
2999 'filepath', 'entries', 'original_url', 'playlist_autonumber',
3001 empty_values
= (None, {}, [], set(), tuple())
3002 reject
= lambda k
, v
: k
not in keep_keys
and (
3003 k
.startswith('_') or k
in remove_keys
or v
in empty_values
)
3005 reject
= lambda k
, v
: k
in remove_keys
3006 filter_fn
= lambda obj
: (
3007 list(map(filter_fn
, obj
)) if isinstance(obj
, (LazyList
, list, tuple, set))
3008 else obj
if not isinstance(obj
, dict)
3009 else dict((k
, filter_fn(v
)) for k
, v
in obj
.items() if not reject(k
, v
)))
3010 return filter_fn(info_dict
)
3013 def filter_requested_info(info_dict
, actually_filter
=True):
3014 ''' Alias of sanitize_info for backward compatibility '''
3015 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3017 def run_pp(self
, pp
, infodict
):
3018 files_to_delete
= []
3019 if '__files_to_move' not in infodict
:
3020 infodict
['__files_to_move'] = {}
3022 files_to_delete
, infodict
= pp
.run(infodict
)
3023 except PostProcessingError
as e
:
3024 # Must be True and not 'only_download'
3025 if self
.params
.get('ignoreerrors') is True:
3026 self
.report_error(e
)
3030 if not files_to_delete
:
3032 if self
.params
.get('keepvideo', False):
3033 for f
in files_to_delete
:
3034 infodict
['__files_to_move'].setdefault(f
, '')
3036 for old_filename
in set(files_to_delete
):
3037 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
3039 os
.remove(encodeFilename(old_filename
))
3040 except (IOError, OSError):
3041 self
.report_warning('Unable to remove downloaded original file')
3042 if old_filename
in infodict
['__files_to_move']:
3043 del infodict
['__files_to_move'][old_filename
]
3047 def post_extract(info_dict
):
3048 def actual_post_extract(info_dict
):
3049 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3050 for video_dict
in info_dict
.get('entries', {}):
3051 actual_post_extract(video_dict
or {})
3054 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
3055 extra
= post_extractor().items()
3056 info_dict
.update(extra
)
3057 info_dict
.pop('__post_extractor', None)
3059 original_infodict
= info_dict
.get('__original_infodict') or {}
3060 original_infodict
.update(extra
)
3061 original_infodict
.pop('__post_extractor', None)
3063 actual_post_extract(info_dict
or {})
3065 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3066 info
= dict(ie_info
)
3067 info
['__files_to_move'] = files_to_move
or {}
3068 for pp
in self
._pps
[key
]:
3069 info
= self
.run_pp(pp
, info
)
3070 return info
, info
.pop('__files_to_move', None)
3072 def post_process(self
, filename
, ie_info
, files_to_move
=None):
3073 """Run all the postprocessors on the given file."""
3074 info
= dict(ie_info
)
3075 info
['filepath'] = filename
3076 info
['__files_to_move'] = files_to_move
or {}
3078 for pp
in ie_info
.get('__postprocessors', []) + self
._pps
['post_process']:
3079 info
= self
.run_pp(pp
, info
)
3080 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3081 del info
['__files_to_move']
3082 for pp
in self
._pps
['after_move']:
3083 info
= self
.run_pp(pp
, info
)
3086 def _make_archive_id(self
, info_dict
):
3087 video_id
= info_dict
.get('id')
3090 # Future-proof against any change in case
3091 # and backwards compatibility with prior versions
3092 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3093 if extractor
is None:
3094 url
= str_or_none(info_dict
.get('url'))
3097 # Try to find matching extractor for the URL and take its ie_key
3098 for ie_key
, ie
in self
._ies
.items():
3099 if ie
.suitable(url
):
3104 return '%s %s' % (extractor
.lower(), video_id
)
3106 def in_download_archive(self
, info_dict
):
3107 fn
= self
.params
.get('download_archive')
3111 vid_id
= self
._make
_archive
_id
(info_dict
)
3113 return False # Incomplete video information
3115 return vid_id
in self
.archive
3117 def record_download_archive(self
, info_dict
):
3118 fn
= self
.params
.get('download_archive')
3121 vid_id
= self
._make
_archive
_id
(info_dict
)
3123 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3124 archive_file
.write(vid_id
+ '\n')
3125 self
.archive
.add(vid_id
)
3128 def format_resolution(format
, default
='unknown'):
3129 is_images
= format
.get('vcodec') == 'none' and format
.get('acodec') == 'none'
3130 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3132 if format
.get('resolution') is not None:
3133 return format
['resolution']
3134 if format
.get('width') and format
.get('height'):
3135 res
= '%dx%d' % (format
['width'], format
['height'])
3136 elif format
.get('height'):
3137 res
= '%sp' % format
['height']
3138 elif format
.get('width'):
3139 res
= '%dx?' % format
['width']
3144 return f
'{res} images' if is_images
else res
3146 def _format_note(self
, fdict
):
3148 if fdict
.get('ext') in ['f4f', 'f4m']:
3149 res
+= '(unsupported) '
3150 if fdict
.get('language'):
3153 res
+= '[%s] ' % fdict
['language']
3154 if fdict
.get('format_note') is not None:
3155 res
+= fdict
['format_note'] + ' '
3156 if fdict
.get('tbr') is not None:
3157 res
+= '%4dk ' % fdict
['tbr']
3158 if fdict
.get('container') is not None:
3161 res
+= '%s container' % fdict
['container']
3162 if (fdict
.get('vcodec') is not None
3163 and fdict
.get('vcodec') != 'none'):
3166 res
+= fdict
['vcodec']
3167 if fdict
.get('vbr') is not None:
3169 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3171 if fdict
.get('vbr') is not None:
3172 res
+= '%4dk' % fdict
['vbr']
3173 if fdict
.get('fps') is not None:
3176 res
+= '%sfps' % fdict
['fps']
3177 if fdict
.get('acodec') is not None:
3180 if fdict
['acodec'] == 'none':
3183 res
+= '%-5s' % fdict
['acodec']
3184 elif fdict
.get('abr') is not None:
3188 if fdict
.get('abr') is not None:
3189 res
+= '@%3dk' % fdict
['abr']
3190 if fdict
.get('asr') is not None:
3191 res
+= ' (%5dHz)' % fdict
['asr']
3192 if fdict
.get('filesize') is not None:
3195 res
+= format_bytes(fdict
['filesize'])
3196 elif fdict
.get('filesize_approx') is not None:
3199 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3202 def _list_format_headers(self
, *headers
):
3203 if self
.params
.get('listformats_table', True) is not False:
3204 return [self
._format
_screen
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3207 def list_formats(self
, info_dict
):
3208 formats
= info_dict
.get('formats', [info_dict
])
3209 new_format
= self
.params
.get('listformats_table', True) is not False
3211 tbr_digits
= number_of_digits(max(f
.get('tbr') or 0 for f
in formats
))
3212 vbr_digits
= number_of_digits(max(f
.get('vbr') or 0 for f
in formats
))
3213 abr_digits
= number_of_digits(max(f
.get('abr') or 0 for f
in formats
))
3214 delim
= self
._format
_screen
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3217 self
._format
_screen
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3218 format_field(f
, 'ext'),
3219 self
.format_resolution(f
),
3220 format_field(f
, 'fps', '%3d'),
3221 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3223 format_field(f
, 'filesize', ' %s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~%s', func
=format_bytes
),
3224 format_field(f
, 'tbr', f
'%{tbr_digits}dk'),
3225 shorten_protocol_name(f
.get('protocol', '').replace("native", "n")),
3227 format_field(f
, 'vcodec', default
='unknown').replace('none', ''),
3228 format_field(f
, 'vbr', f
'%{vbr_digits}dk'),
3229 format_field(f
, 'acodec', default
='unknown').replace('none', ''),
3230 format_field(f
, 'abr', f
'%{abr_digits}dk'),
3231 format_field(f
, 'asr', '%5dHz'),
3233 self
._format
_screen
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3234 format_field(f
, 'language', '[%s]'),
3235 format_field(f
, 'format_note'),
3236 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3238 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3239 header_line
= self
._list
_format
_headers
(
3240 'ID', 'EXT', 'RESOLUTION', 'FPS', 'HDR', delim
, ' FILESIZE', ' TBR', 'PROTO',
3241 delim
, 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO')
3245 format_field(f
, 'format_id'),
3246 format_field(f
, 'ext'),
3247 self
.format_resolution(f
),
3248 self
._format
_note
(f
)]
3250 if f
.get('preference') is None or f
['preference'] >= -1000]
3251 header_line
= ['format code', 'extension', 'resolution', 'note']
3254 '[info] Available formats for %s:' % info_dict
['id'])
3255 self
.to_stdout(render_table(
3257 extraGap
=(0 if new_format
else 1),
3258 hideEmpty
=new_format
,
3259 delim
=new_format
and self
._format
_screen
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True)))
3261 def list_thumbnails(self
, info_dict
):
3262 thumbnails
= list(info_dict
.get('thumbnails'))
3264 self
.to_screen('[info] No thumbnails present for %s' % info_dict
['id'])
3268 '[info] Thumbnails for %s:' % info_dict
['id'])
3269 self
.to_stdout(render_table(
3270 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3271 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
]))
3273 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3275 self
.to_screen('%s has no %s' % (video_id
, name
))
3278 'Available %s for %s:' % (name
, video_id
))
3280 def _row(lang
, formats
):
3281 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3282 if len(set(names
)) == 1:
3283 names
= [] if names
[0] == 'unknown' else names
[:1]
3284 return [lang
, ', '.join(names
), ', '.join(exts
)]
3286 self
.to_stdout(render_table(
3287 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3288 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3291 def urlopen(self
, req
):
3292 """ Start an HTTP download """
3293 if isinstance(req
, compat_basestring
):
3294 req
= sanitized_Request(req
)
3295 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3297 def print_debug_header(self
):
3298 if not self
.params
.get('verbose'):
3301 def get_encoding(stream
):
3302 ret
= getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__)
3303 if not supports_terminal_sequences(stream
):
3307 encoding_str
= 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3308 locale
.getpreferredencoding(),
3309 sys
.getfilesystemencoding(),
3310 get_encoding(self
._screen
_file
), get_encoding(self
._err
_file
),
3311 self
.get_encoding())
3313 logger
= self
.params
.get('logger')
3315 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3316 write_debug(encoding_str
)
3318 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3319 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3321 source
= detect_variant()
3322 write_debug('yt-dlp version %s%s' % (__version__
, '' if source
== 'unknown' else f
' ({source})'))
3323 if not _LAZY_LOADER
:
3324 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3325 write_debug('Lazy loading extractors is forcibly disabled')
3327 write_debug('Lazy loading extractors is disabled')
3328 if plugin_extractors
or plugin_postprocessors
:
3329 write_debug('Plugins: %s' % [
3330 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3331 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3332 if self
.params
.get('compat_opts'):
3333 write_debug('Compatibility options: %s' % ', '.join(self
.params
.get('compat_opts')))
3336 ['git', 'rev-parse', '--short', 'HEAD'],
3337 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3338 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3339 out
, err
= sp
.communicate_or_kill()
3340 out
= out
.decode().strip()
3341 if re
.match('[0-9a-f]+', out
):
3342 write_debug('Git HEAD: %s' % out
)
3349 def python_implementation():
3350 impl_name
= platform
.python_implementation()
3351 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3352 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3355 write_debug('Python version %s (%s %s) - %s' % (
3356 platform
.python_version(),
3357 python_implementation(),
3358 platform
.architecture()[0],
3361 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3362 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3364 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features
)
3366 exe_versions
['rtmpdump'] = rtmpdump_version()
3367 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3368 exe_str
= ', '.join(
3369 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3371 write_debug('exe versions: %s' % exe_str
)
3373 from .downloader
.websocket
import has_websockets
3374 from .postprocessor
.embedthumbnail
import has_mutagen
3375 from .cookies
import SQLITE_AVAILABLE
, KEYRING_AVAILABLE
3377 lib_str
= join_nonempty(
3378 compat_pycrypto_AES
and compat_pycrypto_AES
.__name
__.split('.')[0],
3379 KEYRING_AVAILABLE
and 'keyring',
3380 has_mutagen
and 'mutagen',
3381 SQLITE_AVAILABLE
and 'sqlite',
3382 has_websockets
and 'websockets',
3383 delim
=', ') or 'none'
3384 write_debug('Optional libraries: %s' % lib_str
)
3387 for handler
in self
._opener
.handlers
:
3388 if hasattr(handler
, 'proxies'):
3389 proxy_map
.update(handler
.proxies
)
3390 write_debug(f
'Proxy map: {proxy_map}')
3393 if False and self
.params
.get('call_home'):
3394 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3395 write_debug('Public IP address: %s' % ipaddr
)
3396 latest_version
= self
.urlopen(
3397 'https://yt-dl.org/latest/version').read().decode('utf-8')
3398 if version_tuple(latest_version
) > version_tuple(__version__
):
3399 self
.report_warning(
3400 'You are using an outdated version (newest version: %s)! '
3401 'See https://yt-dl.org/update if you need help updating.' %
3404 def _setup_opener(self
):
3405 timeout_val
= self
.params
.get('socket_timeout')
3406 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3408 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3409 opts_cookiefile
= self
.params
.get('cookiefile')
3410 opts_proxy
= self
.params
.get('proxy')
3412 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3414 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3415 if opts_proxy
is not None:
3416 if opts_proxy
== '':
3419 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3421 proxies
= compat_urllib_request
.getproxies()
3422 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3423 if 'http' in proxies
and 'https' not in proxies
:
3424 proxies
['https'] = proxies
['http']
3425 proxy_handler
= PerRequestProxyHandler(proxies
)
3427 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3428 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3429 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3430 redirect_handler
= YoutubeDLRedirectHandler()
3431 data_handler
= compat_urllib_request_DataHandler()
3433 # When passing our own FileHandler instance, build_opener won't add the
3434 # default FileHandler and allows us to disable the file protocol, which
3435 # can be used for malicious purposes (see
3436 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3437 file_handler
= compat_urllib_request
.FileHandler()
3439 def file_open(*args
, **kwargs
):
3440 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3441 file_handler
.file_open
= file_open
3443 opener
= compat_urllib_request
.build_opener(
3444 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3446 # Delete the default user-agent header, which would otherwise apply in
3447 # cases where our custom HTTP handler doesn't come into play
3448 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3449 opener
.addheaders
= []
3450 self
._opener
= opener
3452 def encode(self
, s
):
3453 if isinstance(s
, bytes):
3454 return s
# Already encoded
3457 return s
.encode(self
.get_encoding())
3458 except UnicodeEncodeError as err
:
3459 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3462 def get_encoding(self
):
3463 encoding
= self
.params
.get('encoding')
3464 if encoding
is None:
3465 encoding
= preferredencoding()
3468 def _write_info_json(self
, label
, ie_result
, infofn
):
3469 ''' Write infojson and returns True = written, False = skip, None = error '''
3470 if not self
.params
.get('writeinfojson'):
3473 self
.write_debug(f
'Skipping writing {label} infojson')
3475 elif not self
._ensure
_dir
_exists
(infofn
):
3477 elif not self
.params
.get('overwrites', True) and os
.path
.exists(infofn
):
3478 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3480 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3482 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3483 except (OSError, IOError):
3484 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3488 def _write_description(self
, label
, ie_result
, descfn
):
3489 ''' Write description and returns True = written, False = skip, None = error '''
3490 if not self
.params
.get('writedescription'):
3493 self
.write_debug(f
'Skipping writing {label} description')
3495 elif not self
._ensure
_dir
_exists
(descfn
):
3497 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3498 self
.to_screen(f
'[info] {label.title()} description is already present')
3499 elif ie_result
.get('description') is None:
3500 self
.report_warning(f
'There\'s no {label} description to write')
3504 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3505 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3506 descfile
.write(ie_result
['description'])
3507 except (OSError, IOError):
3508 self
.report_error(f
'Cannot write {label} description file {descfn}')
3512 def _write_subtitles(self
, info_dict
, filename
):
3513 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3515 subtitles
= info_dict
.get('requested_subtitles')
3516 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3517 # subtitles download errors are already managed as troubles in relevant IE
3518 # that way it will silently go on when used with unsupporting IE
3521 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3522 if not sub_filename_base
:
3523 self
.to_screen('[info] Skipping writing video subtitles')
3525 for sub_lang
, sub_info
in subtitles
.items():
3526 sub_format
= sub_info
['ext']
3527 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3528 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3529 if not self
.params
.get('overwrites', True) and os
.path
.exists(sub_filename
):
3530 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3531 sub_info
['filepath'] = sub_filename
3532 ret
.append((sub_filename
, sub_filename_final
))
3535 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3536 if sub_info
.get('data') is not None:
3538 # Use newline='' to prevent conversion of newline characters
3539 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3540 with io
.open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3541 subfile
.write(sub_info
['data'])
3542 sub_info
['filepath'] = sub_filename
3543 ret
.append((sub_filename
, sub_filename_final
))
3545 except (OSError, IOError):
3546 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3550 sub_copy
= sub_info
.copy()
3551 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3552 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3553 sub_info
['filepath'] = sub_filename
3554 ret
.append((sub_filename
, sub_filename_final
))
3555 except (ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3556 self
.report_warning(f
'Unable to download video subtitles for {sub_lang!r}: {err}')
3560 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3561 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3562 write_all
= self
.params
.get('write_all_thumbnails', False)
3563 thumbnails
, ret
= [], []
3564 if write_all
or self
.params
.get('writethumbnail', False):
3565 thumbnails
= info_dict
.get('thumbnails') or []
3566 multiple
= write_all
and len(thumbnails
) > 1
3568 if thumb_filename_base
is None:
3569 thumb_filename_base
= filename
3570 if thumbnails
and not thumb_filename_base
:
3571 self
.write_debug(f
'Skipping writing {label} thumbnail')
3574 for t
in thumbnails
[::-1]:
3575 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3576 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3577 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3578 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3580 if not self
.params
.get('overwrites', True) and os
.path
.exists(thumb_filename
):
3581 ret
.append((thumb_filename
, thumb_filename_final
))
3582 t
['filepath'] = thumb_filename
3583 self
.to_screen('[info] %s is already present' % (
3584 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3586 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3588 uf
= self
.urlopen(t
['url'])
3589 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3590 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3591 shutil
.copyfileobj(uf
, thumbf
)
3592 ret
.append((thumb_filename
, thumb_filename_final
))
3593 t
['filepath'] = thumb_filename
3594 except network_exceptions
as err
:
3595 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3596 if ret
and not write_all
: