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 You can also pass a function. The function takes 'ctx' as
215 argument and returns the formats to download.
216 See "build_format_selector" for an implementation
217 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
218 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
219 extracting metadata even if the video is not actually
220 available for download (experimental)
221 format_sort: A list of fields by which to sort the video formats.
222 See "Sorting Formats" for more details.
223 format_sort_force: Force the given format_sort. see "Sorting Formats"
225 allow_multiple_video_streams: Allow multiple video streams to be merged
227 allow_multiple_audio_streams: Allow multiple audio streams to be merged
229 check_formats Whether to test if the formats are downloadable.
230 Can be True (check all), False (check none),
231 'selected' (check selected formats),
232 or None (check only if requested by extractor)
233 paths: Dictionary of output paths. The allowed keys are 'home'
234 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
235 outtmpl: Dictionary of templates for output names. Allowed keys
236 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
237 For compatibility with youtube-dl, a single string can also be used
238 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
239 restrictfilenames: Do not allow "&" and spaces in file names
240 trim_file_name: Limit length of filename (extension excluded)
241 windowsfilenames: Force the filenames to be windows compatible
242 ignoreerrors: Do not stop on download/postprocessing errors.
243 Can be 'only_download' to ignore only download errors.
244 Default is 'only_download' for CLI, but False for API
245 skip_playlist_after_errors: Number of allowed failures until the rest of
246 the playlist is skipped
247 force_generic_extractor: Force downloader to use the generic extractor
248 overwrites: Overwrite all video and metadata files if True,
249 overwrite only non-video files if None
250 and don't overwrite any file if False
251 For compatibility with youtube-dl,
252 "nooverwrites" may also be used instead
253 playliststart: Playlist item to start at.
254 playlistend: Playlist item to end at.
255 playlist_items: Specific indices of playlist to download.
256 playlistreverse: Download playlist items in reverse order.
257 playlistrandom: Download playlist items in random order.
258 matchtitle: Download only matching titles.
259 rejecttitle: Reject downloads for matching titles.
260 logger: Log messages to a logging.Logger instance.
261 logtostderr: Log messages to stderr instead of stdout.
262 consoletitle: Display progress in console window's titlebar.
263 writedescription: Write the video description to a .description file
264 writeinfojson: Write the video description to a .info.json file
265 clean_infojson: Remove private fields from the infojson
266 getcomments: Extract video comments. This will not be written to disk
267 unless writeinfojson is also given
268 writeannotations: Write the video annotations to a .annotations.xml file
269 writethumbnail: Write the thumbnail image to a file
270 allow_playlist_files: Whether to write playlists' description, infojson etc
271 also to disk when using the 'write*' options
272 write_all_thumbnails: Write all thumbnail formats to files
273 writelink: Write an internet shortcut file, depending on the
274 current platform (.url/.webloc/.desktop)
275 writeurllink: Write a Windows internet shortcut file (.url)
276 writewebloclink: Write a macOS internet shortcut file (.webloc)
277 writedesktoplink: Write a Linux internet shortcut file (.desktop)
278 writesubtitles: Write the video subtitles to a file
279 writeautomaticsub: Write the automatically generated subtitles to a file
280 allsubtitles: Deprecated - Use subtitleslangs = ['all']
281 Downloads all the subtitles of the video
282 (requires writesubtitles or writeautomaticsub)
283 listsubtitles: Lists all available subtitles for the video
284 subtitlesformat: The format code for subtitles
285 subtitleslangs: List of languages of the subtitles to download (can be regex).
286 The list may contain "all" to refer to all the available
287 subtitles. The language can be prefixed with a "-" to
288 exclude it from the requested languages. Eg: ['all', '-live_chat']
289 keepvideo: Keep the video file after post-processing
290 daterange: A DateRange object, download only if the upload_date is in the range.
291 skip_download: Skip the actual download of the video file
292 cachedir: Location of the cache files in the filesystem.
293 False to disable filesystem cache.
294 noplaylist: Download single video instead of a playlist if in doubt.
295 age_limit: An integer representing the user's age in years.
296 Unsuitable videos for the given age are skipped.
297 min_views: An integer representing the minimum view count the video
298 must have in order to not be skipped.
299 Videos without view count information are always
300 downloaded. None for no limit.
301 max_views: An integer representing the maximum view count.
302 Videos that are more popular than that are not
304 Videos without view count information are always
305 downloaded. None for no limit.
306 download_archive: File name of a file where all downloads are recorded.
307 Videos already present in the file are not downloaded
309 break_on_existing: Stop the download process after attempting to download a
310 file that is in the archive.
311 break_on_reject: Stop the download process when encountering a video that
312 has been filtered out.
313 cookiefile: File name where cookies should be read from and dumped to
314 cookiesfrombrowser: A tuple containing the name of the browser and the profile
315 name/path from where cookies are loaded.
316 Eg: ('chrome', ) or ('vivaldi', 'default')
317 nocheckcertificate:Do not verify SSL certificates
318 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
319 At the moment, this is only supported by YouTube.
320 proxy: URL of the proxy server to use
321 geo_verification_proxy: URL of the proxy to use for IP address verification
322 on geo-restricted sites.
323 socket_timeout: Time to wait for unresponsive hosts, in seconds
324 bidi_workaround: Work around buggy terminals without bidirectional text
325 support, using fridibi
326 debug_printtraffic:Print out sent and received HTTP traffic
327 include_ads: Download ads as well
328 default_search: Prepend this string if an input url is not valid.
329 'auto' for elaborate guessing
330 encoding: Use this encoding instead of the system-specified.
331 extract_flat: Do not resolve URLs, return the immediate result.
332 Pass in 'in_playlist' to only show this behavior for
334 postprocessors: A list of dictionaries, each with an entry
335 * key: The name of the postprocessor. See
336 yt_dlp/postprocessor/__init__.py for a list.
337 * when: When to run the postprocessor. Can be one of
338 pre_process|before_dl|post_process|after_move.
339 Assumed to be 'post_process' if not given
340 post_hooks: Deprecated - Register a custom postprocessor instead
341 A list of functions that get called as the final step
342 for each video file, after all postprocessors have been
343 called. The filename will be passed as the only argument.
344 progress_hooks: A list of functions that get called on download
345 progress, with a dictionary with the entries
346 * status: One of "downloading", "error", or "finished".
347 Check this first and ignore unknown values.
348 * info_dict: The extracted info_dict
350 If status is one of "downloading", or "finished", the
351 following properties may also be present:
352 * filename: The final filename (always present)
353 * tmpfilename: The filename we're currently writing to
354 * downloaded_bytes: Bytes on disk
355 * total_bytes: Size of the whole file, None if unknown
356 * total_bytes_estimate: Guess of the eventual file size,
358 * elapsed: The number of seconds since download started.
359 * eta: The estimated time in seconds, None if unknown
360 * speed: The download speed in bytes/second, None if
362 * fragment_index: The counter of the currently
363 downloaded video fragment.
364 * fragment_count: The number of fragments (= individual
365 files that will be merged)
367 Progress hooks are guaranteed to be called at least once
368 (with status "finished") if the download is successful.
369 postprocessor_hooks: A list of functions that get called on postprocessing
370 progress, with a dictionary with the entries
371 * status: One of "started", "processing", or "finished".
372 Check this first and ignore unknown values.
373 * postprocessor: Name of the postprocessor
374 * info_dict: The extracted info_dict
376 Progress hooks are guaranteed to be called at least twice
377 (with status "started" and "finished") if the processing is successful.
378 merge_output_format: Extension to use when merging formats.
379 final_ext: Expected final extension; used to detect when the file was
380 already downloaded and converted
381 fixup: Automatically correct known faults of the file.
383 - "never": do nothing
384 - "warn": only emit a warning
385 - "detect_or_warn": check whether we can do anything
386 about it, warn otherwise (default)
387 source_address: Client-side IP address to bind to.
388 call_home: Boolean, true iff we are allowed to contact the
389 yt-dlp servers for debugging. (BROKEN)
390 sleep_interval_requests: Number of seconds to sleep between requests
392 sleep_interval: Number of seconds to sleep before each download when
393 used alone or a lower bound of a range for randomized
394 sleep before each download (minimum possible number
395 of seconds to sleep) when used along with
397 max_sleep_interval:Upper bound of a range for randomized sleep before each
398 download (maximum possible number of seconds to sleep).
399 Must only be used along with sleep_interval.
400 Actual sleep time will be a random float from range
401 [sleep_interval; max_sleep_interval].
402 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
403 listformats: Print an overview of available video formats and exit.
404 list_thumbnails: Print a table of all thumbnails and exit.
405 match_filter: A function that gets called with the info_dict of
407 If it returns a message, the video is ignored.
408 If it returns None, the video is downloaded.
409 match_filter_func in utils.py is one example for this.
410 no_color: Do not emit color codes in output.
411 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
414 Two-letter ISO 3166-2 country code that will be used for
415 explicit geographic restriction bypassing via faking
416 X-Forwarded-For HTTP header
418 IP range in CIDR notation that will be used similarly to
421 The following options determine which downloader is picked:
422 external_downloader: A dictionary of protocol keys and the executable of the
423 external downloader to use for it. The allowed protocols
424 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
425 Set the value to 'native' to use the native downloader
426 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
427 or {'m3u8': 'ffmpeg'} instead.
428 Use the native HLS downloader instead of ffmpeg/avconv
429 if True, otherwise use ffmpeg/avconv if False, otherwise
430 use downloader suggested by extractor if None.
431 compat_opts: Compatibility options. See "Differences in default behavior".
432 The following options do not work when used through the API:
433 filename, abort-on-error, multistreams, no-live-chat, format-sort
434 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
435 Refer __init__.py for their implementation
436 progress_template: Dictionary of templates for progress outputs.
437 Allowed keys are 'download', 'postprocess',
438 'download-title' (console title) and 'postprocess-title'.
439 The template is mapped on a dictionary with keys 'progress' and 'info'
441 The following parameters are not used by YoutubeDL itself, they are used by
442 the downloader (see yt_dlp/downloader/common.py):
443 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
444 max_filesize, test, noresizebuffer, retries, fragment_retries, continuedl,
445 noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
446 external_downloader_args, concurrent_fragment_downloads.
448 The following options are used by the post processors:
449 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
450 otherwise prefer ffmpeg. (avconv support is deprecated)
451 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
452 to the binary or its containing directory.
453 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
454 and a list of additional command-line arguments for the
455 postprocessor/executable. The dict can also have "PP+EXE" keys
456 which are used when the given exe is used by the given PP.
457 Use 'default' as the name for arguments to passed to all PP
458 For compatibility with youtube-dl, a single list of args
461 The following options are used by the extractors:
462 extractor_retries: Number of times to retry for known errors
463 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
464 hls_split_discontinuity: Split HLS playlists to different formats at
465 discontinuities such as ad breaks (default: False)
466 extractor_args: A dictionary of arguments to be passed to the extractors.
467 See "EXTRACTOR ARGUMENTS" for details.
468 Eg: {'youtube': {'skip': ['dash', 'hls']}}
469 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
470 If True (default), DASH manifests and related
471 data will be downloaded and processed by extractor.
472 You can reduce network I/O by disabling it if you don't
473 care about DASH. (only for youtube)
474 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
475 If True (default), HLS manifests and related
476 data will be downloaded and processed by extractor.
477 You can reduce network I/O by disabling it if you don't
478 care about HLS. (only for youtube)
481 _NUMERIC_FIELDS
= set((
482 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
483 'timestamp', 'release_timestamp',
484 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
485 'average_rating', 'comment_count', 'age_limit',
486 'start_time', 'end_time',
487 'chapter_number', 'season_number', 'episode_number',
488 'track_number', 'disc_number', 'release_year',
491 _format_selection_exts
= {
492 'audio': {'m4a', 'mp3', 'ogg', 'aac'}
,
493 'video': {'mp4', 'flv', 'webm', '3gp'}
,
494 'storyboards': {'mhtml'}
,
499 _pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
500 _printed_messages
= set()
501 _first_webpage_request
= True
502 _download_retcode
= None
503 _num_downloads
= None
505 _playlist_urls
= set()
508 def __init__(self
, params
=None, auto_init
=True):
509 """Create a FileDownloader object with the given options.
510 @param auto_init Whether to load the default extractors and print header (if verbose).
511 Set to 'no_verbose_header' to not print the header
516 self
._ies
_instances
= {}
517 self
._pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
518 self
._printed
_messages
= set()
519 self
._first
_webpage
_request
= True
520 self
._post
_hooks
= []
521 self
._progress
_hooks
= []
522 self
._postprocessor
_hooks
= []
523 self
._download
_retcode
= 0
524 self
._num
_downloads
= 0
525 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
526 self
._err
_file
= sys
.stderr
528 self
.cache
= Cache(self
)
530 windows_enable_vt_mode()
531 self
._allow
_colors
= {
532 'screen': not self
.params
.get('no_color') and supports_terminal_sequences(self
._screen
_file
),
533 'err': not self
.params
.get('no_color') and supports_terminal_sequences(self
._err
_file
),
536 if sys
.version_info
< (3, 6):
538 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
540 if self
.params
.get('allow_unplayable_formats'):
542 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
543 'This is a developer option intended for debugging. \n'
544 ' If you experience any issues while using this option, '
545 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
547 def check_deprecated(param
, option
, suggestion
):
548 if self
.params
.get(param
) is not None:
549 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
553 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
554 if self
.params
.get('geo_verification_proxy') is None:
555 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
557 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
558 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
559 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
561 for msg
in self
.params
.get('_warnings', []):
562 self
.report_warning(msg
)
564 if 'list-formats' in self
.params
.get('compat_opts', []):
565 self
.params
['listformats_table'] = False
567 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
568 # nooverwrites was unnecessarily changed to overwrites
569 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
570 # This ensures compatibility with both keys
571 self
.params
['overwrites'] = not self
.params
['nooverwrites']
572 elif self
.params
.get('overwrites') is None:
573 self
.params
.pop('overwrites', None)
575 self
.params
['nooverwrites'] = not self
.params
['overwrites']
577 if params
.get('bidi_workaround', False):
580 master
, slave
= pty
.openpty()
581 width
= compat_get_terminal_size().columns
585 width_args
= ['-w', str(width
)]
587 stdin
=subprocess
.PIPE
,
589 stderr
=self
._err
_file
)
591 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
593 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
594 self
._output
_channel
= os
.fdopen(master
, 'rb')
595 except OSError as ose
:
596 if ose
.errno
== errno
.ENOENT
:
598 'Could not find fribidi executable, ignoring --bidi-workaround. '
599 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
603 if (sys
.platform
!= 'win32'
604 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
605 and not params
.get('restrictfilenames', False)):
606 # Unicode filesystem API will throw errors (#1474, #13027)
608 'Assuming --restrict-filenames since file system encoding '
609 'cannot encode all characters. '
610 'Set the LC_ALL environment variable to fix this.')
611 self
.params
['restrictfilenames'] = True
613 self
.outtmpl_dict
= self
.parse_outtmpl()
615 # Creating format selector here allows us to catch syntax errors before the extraction
616 self
.format_selector
= (
617 None if self
.params
.get('format') is None
618 else self
.params
['format'] if callable(self
.params
['format'])
619 else self
.build_format_selector(self
.params
['format']))
624 if auto_init
!= 'no_verbose_header':
625 self
.print_debug_header()
626 self
.add_default_info_extractors()
628 for pp_def_raw
in self
.params
.get('postprocessors', []):
629 pp_def
= dict(pp_def_raw
)
630 when
= pp_def
.pop('when', 'post_process')
631 pp_class
= get_postprocessor(pp_def
.pop('key'))
632 pp
= pp_class(self
, **compat_kwargs(pp_def
))
633 self
.add_post_processor(pp
, when
=when
)
636 'post_hooks': self
.add_post_hook
,
637 'progress_hooks': self
.add_progress_hook
,
638 'postprocessor_hooks': self
.add_postprocessor_hook
,
640 for opt
, fn
in hooks
.items():
641 for ph
in self
.params
.get(opt
, []):
644 register_socks_protocols()
646 def preload_download_archive(fn
):
647 """Preload the archive, if any is specified"""
650 self
.write_debug(f
'Loading archive file {fn!r}')
652 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
653 for line
in archive_file
:
654 self
.archive
.add(line
.strip())
655 except IOError as ioe
:
656 if ioe
.errno
!= errno
.ENOENT
:
662 preload_download_archive(self
.params
.get('download_archive'))
664 def warn_if_short_id(self
, argv
):
665 # short YouTube ID starting with dash?
667 i
for i
, a
in enumerate(argv
)
668 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
672 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
673 + ['--'] + [argv
[i
] for i
in idxs
]
676 'Long argument string detected. '
677 'Use -- to separate parameters and URLs, like this:\n%s' %
678 args_to_str(correct_argv
))
680 def add_info_extractor(self
, ie
):
681 """Add an InfoExtractor object to the end of the list."""
683 self
._ies
[ie_key
] = ie
684 if not isinstance(ie
, type):
685 self
._ies
_instances
[ie_key
] = ie
686 ie
.set_downloader(self
)
688 def _get_info_extractor_class(self
, ie_key
):
689 ie
= self
._ies
.get(ie_key
)
691 ie
= get_info_extractor(ie_key
)
692 self
.add_info_extractor(ie
)
695 def get_info_extractor(self
, ie_key
):
697 Get an instance of an IE with name ie_key, it will try to get one from
698 the _ies list, if there's no instance it will create a new one and add
699 it to the extractor list.
701 ie
= self
._ies
_instances
.get(ie_key
)
703 ie
= get_info_extractor(ie_key
)()
704 self
.add_info_extractor(ie
)
707 def add_default_info_extractors(self
):
709 Add the InfoExtractors returned by gen_extractors to the end of the list
711 for ie
in gen_extractor_classes():
712 self
.add_info_extractor(ie
)
714 def add_post_processor(self
, pp
, when
='post_process'):
715 """Add a PostProcessor object to the end of the chain."""
716 self
._pps
[when
].append(pp
)
717 pp
.set_downloader(self
)
719 def add_post_hook(self
, ph
):
720 """Add the post hook"""
721 self
._post
_hooks
.append(ph
)
723 def add_progress_hook(self
, ph
):
724 """Add the download progress hook"""
725 self
._progress
_hooks
.append(ph
)
727 def add_postprocessor_hook(self
, ph
):
728 """Add the postprocessing progress hook"""
729 self
._postprocessor
_hooks
.append(ph
)
731 def _bidi_workaround(self
, message
):
732 if not hasattr(self
, '_output_channel'):
735 assert hasattr(self
, '_output_process')
736 assert isinstance(message
, compat_str
)
737 line_count
= message
.count('\n') + 1
738 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
739 self
._output
_process
.stdin
.flush()
740 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
741 for _
in range(line_count
))
742 return res
[:-len('\n')]
744 def _write_string(self
, message
, out
=None, only_once
=False):
746 if message
in self
._printed
_messages
:
748 self
._printed
_messages
.add(message
)
749 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
751 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
752 """Print message to stdout"""
753 if self
.params
.get('logger'):
754 self
.params
['logger'].debug(message
)
755 elif not quiet
or self
.params
.get('verbose'):
757 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
758 self
._err
_file
if quiet
else self
._screen
_file
)
760 def to_stderr(self
, message
, only_once
=False):
761 """Print message to stderr"""
762 assert isinstance(message
, compat_str
)
763 if self
.params
.get('logger'):
764 self
.params
['logger'].error(message
)
766 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._err
_file
, only_once
=only_once
)
768 def to_console_title(self
, message
):
769 if not self
.params
.get('consoletitle', False):
771 if compat_os_name
== 'nt':
772 if ctypes
.windll
.kernel32
.GetConsoleWindow():
773 # c_wchar_p() might not be necessary if `message` is
774 # already of type unicode()
775 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
776 elif 'TERM' in os
.environ
:
777 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
779 def save_console_title(self
):
780 if not self
.params
.get('consoletitle', False):
782 if self
.params
.get('simulate'):
784 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
785 # Save the title on stack
786 self
._write
_string
('\033[22;0t', self
._screen
_file
)
788 def restore_console_title(self
):
789 if not self
.params
.get('consoletitle', False):
791 if self
.params
.get('simulate'):
793 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
794 # Restore the title from stack
795 self
._write
_string
('\033[23;0t', self
._screen
_file
)
798 self
.save_console_title()
801 def __exit__(self
, *args
):
802 self
.restore_console_title()
804 if self
.params
.get('cookiefile') is not None:
805 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
807 def trouble(self
, message
=None, tb
=None):
808 """Determine action to take when a download problem appears.
810 Depending on if the downloader has been configured to ignore
811 download errors or not, this method may throw an exception or
812 not when errors are found, after printing the message.
814 tb, if given, is additional traceback information.
816 if message
is not None:
817 self
.to_stderr(message
)
818 if self
.params
.get('verbose'):
820 if sys
.exc_info()[0]: # if .trouble has been called from an except block
822 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
823 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
824 tb
+= encode_compat_str(traceback
.format_exc())
826 tb_data
= traceback
.format_list(traceback
.extract_stack())
827 tb
= ''.join(tb_data
)
830 if not self
.params
.get('ignoreerrors'):
831 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
832 exc_info
= sys
.exc_info()[1].exc_info
834 exc_info
= sys
.exc_info()
835 raise DownloadError(message
, exc_info
)
836 self
._download
_retcode
= 1
838 def to_screen(self
, message
, skip_eol
=False):
839 """Print message to stdout if not in quiet mode"""
841 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
851 def __format_text(self
, out
, text
, f
, fallback
=None, *, test_encoding
=False):
852 assert out
in ('screen', 'err')
855 handle
= self
._screen
_file
if out
== 'screen' else self
._err
_file
856 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', 'ascii')
857 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
858 if fallback
is not None and text
!= original_text
:
860 if isinstance(f
, self
.Styles
):
862 return format_text(text
, f
) if self
._allow
_colors
[out
] else text
if fallback
is None else fallback
864 def _format_screen(self
, *args
, **kwargs
):
865 return self
.__format
_text
('screen', *args
, **kwargs
)
867 def _format_err(self
, *args
, **kwargs
):
868 return self
.__format
_text
('err', *args
, **kwargs
)
870 def report_warning(self
, message
, only_once
=False):
872 Print the message to stderr, it will be prefixed with 'WARNING:'
873 If stderr is a tty file the 'WARNING:' will be colored
875 if self
.params
.get('logger') is not None:
876 self
.params
['logger'].warning(message
)
878 if self
.params
.get('no_warnings'):
880 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
882 def report_error(self
, message
, tb
=None):
884 Do the same as trouble, but prefixes the message with 'ERROR:', colored
885 in red if stderr is a tty file.
887 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', tb
)
889 def write_debug(self
, message
, only_once
=False):
890 '''Log debug message or Print message to stderr'''
891 if not self
.params
.get('verbose', False):
893 message
= '[debug] %s' % message
894 if self
.params
.get('logger'):
895 self
.params
['logger'].debug(message
)
897 self
.to_stderr(message
, only_once
)
899 def report_file_already_downloaded(self
, file_name
):
900 """Report file has already been fully downloaded."""
902 self
.to_screen('[download] %s has already been downloaded' % file_name
)
903 except UnicodeEncodeError:
904 self
.to_screen('[download] The file has already been downloaded')
906 def report_file_delete(self
, file_name
):
907 """Report that existing file will be deleted."""
909 self
.to_screen('Deleting existing file %s' % file_name
)
910 except UnicodeEncodeError:
911 self
.to_screen('Deleting existing file')
913 def raise_no_formats(self
, info
, forced
=False):
914 has_drm
= info
.get('__has_drm')
915 msg
= 'This video is DRM protected' if has_drm
else 'No video formats found!'
916 expected
= self
.params
.get('ignore_no_formats_error')
917 if forced
or not expected
:
918 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
919 expected
=has_drm
or expected
)
921 self
.report_warning(msg
)
923 def parse_outtmpl(self
):
924 outtmpl_dict
= self
.params
.get('outtmpl', {})
925 if not isinstance(outtmpl_dict
, dict):
926 outtmpl_dict
= {'default': outtmpl_dict}
927 # Remove spaces in the default template
928 if self
.params
.get('restrictfilenames'):
929 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
931 sanitize
= lambda x
: x
932 outtmpl_dict
.update({
933 k
: sanitize(v
) for k
, v
in DEFAULT_OUTTMPL
.items()
934 if outtmpl_dict
.get(k
) is None})
935 for key
, val
in outtmpl_dict
.items():
936 if isinstance(val
, bytes):
938 'Parameter outtmpl is bytes, but should be a unicode string. '
939 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
942 def get_output_path(self
, dir_type
='', filename
=None):
943 paths
= self
.params
.get('paths', {})
944 assert isinstance(paths
, dict)
946 expand_path(paths
.get('home', '').strip()),
947 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
950 # Temporary fix for #4787
951 # 'Treat' all problem characters by passing filename through preferredencoding
952 # to workaround encoding issues with subprocess on python2 @ Windows
953 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
954 path
= encodeFilename(path
, True).decode(preferredencoding())
955 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
958 def _outtmpl_expandpath(outtmpl
):
959 # expand_path translates '%%' into '%' and '$$' into '$'
960 # correspondingly that is not what we want since we need to keep
961 # '%%' intact for template dict substitution step. Working around
962 # with boundary-alike separator hack.
963 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
964 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
966 # outtmpl should be expand_path'ed before template dict substitution
967 # because meta fields may contain env variables we don't want to
968 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
969 # title "Hello $PATH", we don't want `$PATH` to be expanded.
970 return expand_path(outtmpl
).replace(sep
, '')
973 def escape_outtmpl(outtmpl
):
974 ''' Escape any remaining strings like %s, %abc% etc. '''
976 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
977 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
981 def validate_outtmpl(cls
, outtmpl
):
982 ''' @return None or Exception object '''
984 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljqBU]'),
985 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
986 cls
._outtmpl
_expandpath
(outtmpl
))
988 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
990 except ValueError as err
:
994 def _copy_infodict(info_dict
):
995 info_dict
= dict(info_dict
)
996 for key
in ('__original_infodict', '__postprocessors'):
997 info_dict
.pop(key
, None)
1000 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=None):
1001 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict """
1002 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1004 info_dict
= self
._copy
_infodict
(info_dict
)
1005 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1006 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1007 if info_dict
.get('duration', None) is not None
1009 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
1010 if info_dict
.get('resolution') is None:
1011 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1013 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1014 # of %(field)s to %(field)0Nd for backward compatibility
1015 field_size_compat_map
= {
1016 'playlist_index': number_of_digits(info_dict
.get('_last_playlist_index') or 0),
1017 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1018 'autonumber': self
.params
.get('autonumber_size') or 5,
1022 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljqBU]'))
1027 # Field is of the form key1.key2...
1028 # where keys (except first) can be string, int or slice
1029 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1030 MATH_FIELD_RE
= r
'''{field}|{num}'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
1031 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1032 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
1035 (?P<maths>(?:{math_op}{math_field})*)
1036 (?:>(?P<strf_format>.+?))?
1037 (?P<alternate>(?<!\\),[^|)]+)?
1038 (?:\|(?P<default>.*?))?
1039 $'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
1041 def _traverse_infodict(k
):
1045 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1047 def get_value(mdict
):
1049 value
= _traverse_infodict(mdict
['fields'])
1052 value
= float_or_none(value
)
1053 if value
is not None:
1056 offset_key
= mdict
['maths']
1058 value
= float_or_none(value
)
1062 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1063 offset_key
).group(0)
1064 offset_key
= offset_key
[len(item
):]
1065 if operator
is None:
1066 operator
= MATH_FUNCTIONS
[item
]
1068 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1069 offset
= float_or_none(item
)
1071 offset
= float_or_none(_traverse_infodict(item
))
1073 value
= operator(value
, multiplier
* offset
)
1074 except (TypeError, ZeroDivisionError):
1077 # Datetime formatting
1078 if mdict
['strf_format']:
1079 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1083 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1085 def _dumpjson_default(obj
):
1086 if isinstance(obj
, (set, LazyList
)):
1088 raise TypeError(f
'Object of type {type(obj).__name__} is not JSON serializable')
1090 def create_key(outer_mobj
):
1091 if not outer_mobj
.group('has_key'):
1092 return outer_mobj
.group(0)
1093 key
= outer_mobj
.group('key')
1094 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1095 initial_field
= mobj
.group('fields').split('.')[-1] if mobj
else ''
1096 value
, default
= None, na
1098 mobj
= mobj
.groupdict()
1099 default
= mobj
['default'] if mobj
['default'] is not None else default
1100 value
= get_value(mobj
)
1101 if value
is None and mobj
['alternate']:
1102 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['alternate'][1:])
1106 fmt
= outer_mobj
.group('format')
1107 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1108 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
1110 value
= default
if value
is None else value
1112 flags
= outer_mobj
.group('conversion') or ''
1113 str_fmt
= f
'{fmt[:-1]}s'
1114 if fmt
[-1] == 'l': # list
1115 delim
= '\n' if '#' in flags
else ', '
1116 value
, fmt
= delim
.join(variadic(value
)), str_fmt
1117 elif fmt
[-1] == 'j': # json
1118 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1119 elif fmt
[-1] == 'q': # quoted
1120 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1121 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1122 elif fmt
[-1] == 'B': # bytes
1123 value
= f
'%{str_fmt}'.encode('utf-8') % str(value
).encode('utf-8')
1124 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1125 elif fmt
[-1] == 'U': # unicode normalized
1126 value
, fmt
= unicodedata
.normalize(
1127 # "+" = compatibility equivalence, "#" = NFD
1128 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1130 elif fmt
[-1] == 'c':
1132 value
= str(value
)[0]
1135 elif fmt
[-1] not in 'rs': # numeric
1136 value
= float_or_none(value
)
1138 value
, fmt
= default
, 's'
1142 # If value is an object, sanitize might convert it to a string
1143 # So we convert it to repr first
1144 value
, fmt
= repr(value
), str_fmt
1145 if fmt
[-1] in 'csr':
1146 value
= sanitize(initial_field
, value
)
1148 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1149 TMPL_DICT
[key
] = value
1150 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1152 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1154 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1155 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1156 return self
.escape_outtmpl(outtmpl
) % info_dict
1158 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
1160 sanitize
= lambda k
, v
: sanitize_filename(
1162 restricted
=self
.params
.get('restrictfilenames'),
1163 is_id
=(k
== 'id' or k
.endswith('_id')))
1164 outtmpl
= self
._outtmpl
_expandpath
(self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default']))
1165 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, sanitize
)
1167 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
1168 if filename
and force_ext
is not None:
1169 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1171 # https://github.com/blackjack4494/youtube-dlc/issues/85
1172 trim_file_name
= self
.params
.get('trim_file_name', False)
1174 fn_groups
= filename
.rsplit('.')
1177 if len(fn_groups
) > 2:
1178 sub_ext
= fn_groups
[-2]
1179 filename
= join_nonempty(fn_groups
[0][:trim_file_name
], sub_ext
, ext
, delim
='.')
1182 except ValueError as err
:
1183 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1186 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
1187 """Generate the output filename."""
1189 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
1190 if not filename
and dir_type
not in ('', 'temp'):
1194 if not self
.params
.get('paths'):
1196 elif filename
== '-':
1197 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1198 elif os
.path
.isabs(filename
):
1199 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1200 if filename
== '-' or not filename
:
1203 return self
.get_output_path(dir_type
, filename
)
1205 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1206 """ Returns None if the file should be downloaded """
1208 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1211 if 'title' in info_dict
:
1212 # This can happen when we're just evaluating the playlist
1213 title
= info_dict
['title']
1214 matchtitle
= self
.params
.get('matchtitle', False)
1216 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1217 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1218 rejecttitle
= self
.params
.get('rejecttitle', False)
1220 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1221 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1222 date
= info_dict
.get('upload_date')
1223 if date
is not None:
1224 dateRange
= self
.params
.get('daterange', DateRange())
1225 if date
not in dateRange
:
1226 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1227 view_count
= info_dict
.get('view_count')
1228 if view_count
is not None:
1229 min_views
= self
.params
.get('min_views')
1230 if min_views
is not None and view_count
< min_views
:
1231 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1232 max_views
= self
.params
.get('max_views')
1233 if max_views
is not None and view_count
> max_views
:
1234 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1235 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1236 return 'Skipping "%s" because it is age restricted' % video_title
1238 match_filter
= self
.params
.get('match_filter')
1239 if match_filter
is not None:
1241 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1243 # For backward compatibility
1244 ret
= None if incomplete
else match_filter(info_dict
)
1249 if self
.in_download_archive(info_dict
):
1250 reason
= '%s has already been recorded in the archive' % video_title
1251 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1253 reason
= check_filter()
1254 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1255 if reason
is not None:
1257 self
.to_screen('[download] ' + reason
)
1258 if self
.params
.get(break_opt
, False):
1263 def add_extra_info(info_dict
, extra_info
):
1264 '''Set the keys from extra_info in info dict if they are missing'''
1265 for key
, value
in extra_info
.items():
1266 info_dict
.setdefault(key
, value
)
1268 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1269 process
=True, force_generic_extractor
=False):
1271 Return a list with a dictionary for each video extracted.
1274 url -- URL to extract
1277 download -- whether to download videos during extraction
1278 ie_key -- extractor key hint
1279 extra_info -- dictionary containing the extra values to add to each result
1280 process -- whether to resolve all unresolved references (URLs, playlist items),
1281 must be True for download to work.
1282 force_generic_extractor -- force using the generic extractor
1285 if extra_info
is None:
1288 if not ie_key
and force_generic_extractor
:
1292 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1296 for ie_key
, ie
in ies
.items():
1297 if not ie
.suitable(url
):
1300 if not ie
.working():
1301 self
.report_warning('The program functionality for this site has been marked as broken, '
1302 'and will probably not work.')
1304 temp_id
= ie
.get_temp_id(url
)
1305 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1306 self
.to_screen("[%s] %s: has already been recorded in archive" % (
1309 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1311 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1313 def __handle_extraction_exceptions(func
):
1314 @functools.wraps(func
)
1315 def wrapper(self
, *args
, **kwargs
):
1317 return func(self
, *args
, **kwargs
)
1318 except GeoRestrictedError
as e
:
1321 msg
+= '\nThis video is available in %s.' % ', '.join(
1322 map(ISO3166Utils
.short2full
, e
.countries
))
1323 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1324 self
.report_error(msg
)
1325 except ExtractorError
as e
: # An error we somewhat expected
1326 self
.report_error(compat_str(e
), e
.format_traceback())
1327 except ThrottledDownload
as e
:
1328 self
.to_stderr('\r')
1329 self
.report_warning(f
'{e}; Re-extracting data')
1330 return wrapper(self
, *args
, **kwargs
)
1331 except (DownloadCancelled
, LazyList
.IndexError):
1333 except Exception as e
:
1334 if self
.params
.get('ignoreerrors'):
1335 self
.report_error(error_to_compat_str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1340 @__handle_extraction_exceptions
1341 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1342 ie_result
= ie
.extract(url
)
1343 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1345 if isinstance(ie_result
, list):
1346 # Backwards compatibility: old IE result format
1348 '_type': 'compat_list',
1349 'entries': ie_result
,
1351 if extra_info
.get('original_url'):
1352 ie_result
.setdefault('original_url', extra_info
['original_url'])
1353 self
.add_default_extra_info(ie_result
, ie
, url
)
1355 return self
.process_ie_result(ie_result
, download
, extra_info
)
1359 def add_default_extra_info(self
, ie_result
, ie
, url
):
1361 self
.add_extra_info(ie_result
, {
1363 'original_url': url
,
1364 'webpage_url_basename': url_basename(url
),
1367 self
.add_extra_info(ie_result
, {
1368 'extractor': ie
.IE_NAME
,
1369 'extractor_key': ie
.ie_key(),
1372 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1374 Take the result of the ie(may be modified) and resolve all unresolved
1375 references (URLs, playlist items).
1377 It will also download the videos if 'download'.
1378 Returns the resolved ie_result.
1380 if extra_info
is None:
1382 result_type
= ie_result
.get('_type', 'video')
1384 if result_type
in ('url', 'url_transparent'):
1385 ie_result
['url'] = sanitize_url(ie_result
['url'])
1386 if ie_result
.get('original_url'):
1387 extra_info
.setdefault('original_url', ie_result
['original_url'])
1389 extract_flat
= self
.params
.get('extract_flat', False)
1390 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1391 or extract_flat
is True):
1392 info_copy
= ie_result
.copy()
1393 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1394 if ie
and not ie_result
.get('id'):
1395 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1396 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1397 self
.add_extra_info(info_copy
, extra_info
)
1398 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1399 if self
.params
.get('force_write_download_archive', False):
1400 self
.record_download_archive(info_copy
)
1403 if result_type
== 'video':
1404 self
.add_extra_info(ie_result
, extra_info
)
1405 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1406 additional_urls
= (ie_result
or {}).get('additional_urls')
1408 # TODO: Improve MetadataParserPP to allow setting a list
1409 if isinstance(additional_urls
, compat_str
):
1410 additional_urls
= [additional_urls
]
1412 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1413 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1414 ie_result
['additional_entries'] = [
1416 url
, download
, extra_info
,
1417 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1418 for url
in additional_urls
1421 elif result_type
== 'url':
1422 # We have to add extra_info to the results because it may be
1423 # contained in a playlist
1424 return self
.extract_info(
1425 ie_result
['url'], download
,
1426 ie_key
=ie_result
.get('ie_key'),
1427 extra_info
=extra_info
)
1428 elif result_type
== 'url_transparent':
1429 # Use the information from the embedding page
1430 info
= self
.extract_info(
1431 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1432 extra_info
=extra_info
, download
=False, process
=False)
1434 # extract_info may return None when ignoreerrors is enabled and
1435 # extraction failed with an error, don't crash and return early
1440 force_properties
= dict(
1441 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1442 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1443 if f
in force_properties
:
1444 del force_properties
[f
]
1445 new_result
= info
.copy()
1446 new_result
.update(force_properties
)
1448 # Extracted info may not be a video result (i.e.
1449 # info.get('_type', 'video') != video) but rather an url or
1450 # url_transparent. In such cases outer metadata (from ie_result)
1451 # should be propagated to inner one (info). For this to happen
1452 # _type of info should be overridden with url_transparent. This
1453 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1454 if new_result
.get('_type') == 'url':
1455 new_result
['_type'] = 'url_transparent'
1457 return self
.process_ie_result(
1458 new_result
, download
=download
, extra_info
=extra_info
)
1459 elif result_type
in ('playlist', 'multi_video'):
1460 # Protect from infinite recursion due to recursively nested playlists
1461 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1462 webpage_url
= ie_result
['webpage_url']
1463 if webpage_url
in self
._playlist
_urls
:
1465 '[download] Skipping already downloaded playlist: %s'
1466 % ie_result
.get('title') or ie_result
.get('id'))
1469 self
._playlist
_level
+= 1
1470 self
._playlist
_urls
.add(webpage_url
)
1471 self
._sanitize
_thumbnails
(ie_result
)
1473 return self
.__process
_playlist
(ie_result
, download
)
1475 self
._playlist
_level
-= 1
1476 if not self
._playlist
_level
:
1477 self
._playlist
_urls
.clear()
1478 elif result_type
== 'compat_list':
1479 self
.report_warning(
1480 'Extractor %s returned a compat_list result. '
1481 'It needs to be updated.' % ie_result
.get('extractor'))
1484 self
.add_extra_info(r
, {
1485 'extractor': ie_result
['extractor'],
1486 'webpage_url': ie_result
['webpage_url'],
1487 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1488 'extractor_key': ie_result
['extractor_key'],
1491 ie_result
['entries'] = [
1492 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1493 for r
in ie_result
['entries']
1497 raise Exception('Invalid result type: %s' % result_type
)
1499 def _ensure_dir_exists(self
, path
):
1500 return make_dir(path
, self
.report_error
)
1502 def __process_playlist(self
, ie_result
, download
):
1503 # We process each entry in the playlist
1504 playlist
= ie_result
.get('title') or ie_result
.get('id')
1505 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1507 if 'entries' not in ie_result
:
1508 raise EntryNotInPlaylist('There are no entries')
1510 MissingEntry
= object()
1511 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1512 if incomplete_entries
:
1513 def fill_missing_entries(entries
, indices
):
1514 ret
= [MissingEntry
] * max(indices
)
1515 for i
, entry
in zip(indices
, entries
):
1518 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1520 playlist_results
= []
1522 playliststart
= self
.params
.get('playliststart', 1)
1523 playlistend
= self
.params
.get('playlistend')
1524 # For backwards compatibility, interpret -1 as whole list
1525 if playlistend
== -1:
1528 playlistitems_str
= self
.params
.get('playlist_items')
1529 playlistitems
= None
1530 if playlistitems_str
is not None:
1531 def iter_playlistitems(format
):
1532 for string_segment
in format
.split(','):
1533 if '-' in string_segment
:
1534 start
, end
= string_segment
.split('-')
1535 for item
in range(int(start
), int(end
) + 1):
1538 yield int(string_segment
)
1539 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1541 ie_entries
= ie_result
['entries']
1543 'Downloading %d videos' if not isinstance(ie_entries
, list)
1544 else 'Collected %d videos; downloading %%d of them' % len(ie_entries
))
1546 if isinstance(ie_entries
, list):
1548 return ie_entries
[i
- 1]
1550 if not isinstance(ie_entries
, (PagedList
, LazyList
)):
1551 ie_entries
= LazyList(ie_entries
)
1554 return YoutubeDL
.__handle
_extraction
_exceptions
(
1555 lambda self
, i
: ie_entries
[i
- 1]
1559 items
= playlistitems
if playlistitems
is not None else itertools
.count(playliststart
)
1563 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1567 entry
= get_entry(i
)
1568 if entry
is MissingEntry
:
1569 raise EntryNotInPlaylist()
1570 except (IndexError, EntryNotInPlaylist
):
1571 if incomplete_entries
:
1572 raise EntryNotInPlaylist(f
'Entry {i} cannot be found')
1573 elif not playlistitems
:
1575 entries
.append(entry
)
1577 if entry
is not None:
1578 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1579 except (ExistingVideoReached
, RejectedVideoReached
):
1581 ie_result
['entries'] = entries
1583 # Save playlist_index before re-ordering
1585 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1586 for i
, entry
in enumerate(entries
, 1)
1587 if entry
is not None]
1588 n_entries
= len(entries
)
1590 if not playlistitems
and (playliststart
!= 1 or playlistend
):
1591 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1592 ie_result
['requested_entries'] = playlistitems
1594 _infojson_written
= False
1595 if not self
.params
.get('simulate') and self
.params
.get('allow_playlist_files', True):
1597 'playlist': playlist
,
1598 'playlist_id': ie_result
.get('id'),
1599 'playlist_title': ie_result
.get('title'),
1600 'playlist_uploader': ie_result
.get('uploader'),
1601 'playlist_uploader_id': ie_result
.get('uploader_id'),
1602 'playlist_index': 0,
1603 'n_entries': n_entries
,
1605 ie_copy
.update(dict(ie_result
))
1607 _infojson_written
= self
._write
_info
_json
(
1608 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1609 if _infojson_written
is None:
1611 if self
._write
_description
('playlist', ie_result
,
1612 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1614 # TODO: This should be passed to ThumbnailsConvertor if necessary
1615 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1617 if self
.params
.get('playlistreverse', False):
1618 entries
= entries
[::-1]
1619 if self
.params
.get('playlistrandom', False):
1620 random
.shuffle(entries
)
1622 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1624 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1626 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1627 for i
, entry_tuple
in enumerate(entries
, 1):
1628 playlist_index
, entry
= entry_tuple
1629 if 'playlist-index' in self
.params
.get('compat_opts', []):
1630 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1
1631 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1632 # This __x_forwarded_for_ip thing is a bit ugly but requires
1635 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1637 'n_entries': n_entries
,
1638 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1639 'playlist_index': playlist_index
,
1640 'playlist_autonumber': i
,
1641 'playlist': playlist
,
1642 'playlist_id': ie_result
.get('id'),
1643 'playlist_title': ie_result
.get('title'),
1644 'playlist_uploader': ie_result
.get('uploader'),
1645 'playlist_uploader_id': ie_result
.get('uploader_id'),
1646 'extractor': ie_result
['extractor'],
1647 'webpage_url': ie_result
['webpage_url'],
1648 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1649 'extractor_key': ie_result
['extractor_key'],
1652 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1655 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1656 if not entry_result
:
1658 if failures
>= max_failures
:
1660 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1662 playlist_results
.append(entry_result
)
1663 ie_result
['entries'] = playlist_results
1665 # Write the updated info to json
1666 if _infojson_written
and self
._write
_info
_json
(
1667 'updated playlist', ie_result
,
1668 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1670 self
.to_screen('[download] Finished downloading playlist: %s' % playlist
)
1673 @__handle_extraction_exceptions
1674 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1675 return self
.process_ie_result(
1676 entry
, download
=download
, extra_info
=extra_info
)
1678 def _build_format_filter(self
, filter_spec
):
1679 " Returns a function to filter the formats according to the filter_spec "
1689 operator_rex
= re
.compile(r
'''(?x)\s*
1690 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1691 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1692 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1693 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1694 m
= operator_rex
.fullmatch(filter_spec
)
1697 comparison_value
= int(m
.group('value'))
1699 comparison_value
= parse_filesize(m
.group('value'))
1700 if comparison_value
is None:
1701 comparison_value
= parse_filesize(m
.group('value') + 'B')
1702 if comparison_value
is None:
1704 'Invalid value %r in format specification %r' % (
1705 m
.group('value'), filter_spec
))
1706 op
= OPERATORS
[m
.group('op')]
1711 '^=': lambda attr
, value
: attr
.startswith(value
),
1712 '$=': lambda attr
, value
: attr
.endswith(value
),
1713 '*=': lambda attr
, value
: value
in attr
,
1715 str_operator_rex
= re
.compile(r
'''(?x)\s*
1716 (?P<key>[a-zA-Z0-9._-]+)\s*
1717 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1718 (?P<value>[a-zA-Z0-9._-]+)\s*
1719 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1720 m
= str_operator_rex
.fullmatch(filter_spec
)
1722 comparison_value
= m
.group('value')
1723 str_op
= STR_OPERATORS
[m
.group('op')]
1724 if m
.group('negation'):
1725 op
= lambda attr
, value
: not str_op(attr
, value
)
1730 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1733 actual_value
= f
.get(m
.group('key'))
1734 if actual_value
is None:
1735 return m
.group('none_inclusive')
1736 return op(actual_value
, comparison_value
)
1739 def _check_formats(self
, formats
):
1741 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1742 temp_file
= tempfile
.NamedTemporaryFile(
1743 suffix
='.tmp', delete
=False,
1744 dir=self
.get_output_path('temp') or None)
1747 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1748 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1751 if os
.path
.exists(temp_file
.name
):
1753 os
.remove(temp_file
.name
)
1755 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1759 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1761 def _default_format_spec(self
, info_dict
, download
=True):
1764 merger
= FFmpegMergerPP(self
)
1765 return merger
.available
and merger
.can_merge()
1768 not self
.params
.get('simulate')
1772 or info_dict
.get('is_live', False)
1773 or self
.outtmpl_dict
['default'] == '-'))
1776 or self
.params
.get('allow_multiple_audio_streams', False)
1777 or 'format-spec' in self
.params
.get('compat_opts', []))
1780 'best/bestvideo+bestaudio' if prefer_best
1781 else 'bestvideo*+bestaudio/best' if not compat
1782 else 'bestvideo+bestaudio/best')
1784 def build_format_selector(self
, format_spec
):
1785 def syntax_error(note
, start
):
1787 'Invalid format specification: '
1788 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1789 return SyntaxError(message
)
1791 PICKFIRST
= 'PICKFIRST'
1795 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1797 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1798 'video': self
.params
.get('allow_multiple_video_streams', False)}
1800 check_formats
= self
.params
.get('check_formats') == 'selected'
1802 def _parse_filter(tokens
):
1804 for type, string
, start
, _
, _
in tokens
:
1805 if type == tokenize
.OP
and string
== ']':
1806 return ''.join(filter_parts
)
1808 filter_parts
.append(string
)
1810 def _remove_unused_ops(tokens
):
1811 # Remove operators that we don't use and join them with the surrounding strings
1812 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1813 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1814 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1815 for type, string
, start
, end
, line
in tokens
:
1816 if type == tokenize
.OP
and string
== '[':
1818 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1820 yield type, string
, start
, end
, line
1821 # everything inside brackets will be handled by _parse_filter
1822 for type, string
, start
, end
, line
in tokens
:
1823 yield type, string
, start
, end
, line
1824 if type == tokenize
.OP
and string
== ']':
1826 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1828 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1830 yield type, string
, start
, end
, line
1831 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1833 last_string
= string
1837 last_string
+= string
1839 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1841 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1843 current_selector
= None
1844 for type, string
, start
, _
, _
in tokens
:
1845 # ENCODING is only defined in python 3.x
1846 if type == getattr(tokenize
, 'ENCODING', None):
1848 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1849 current_selector
= FormatSelector(SINGLE
, string
, [])
1850 elif type == tokenize
.OP
:
1852 if not inside_group
:
1853 # ')' will be handled by the parentheses group
1854 tokens
.restore_last_token()
1856 elif inside_merge
and string
in ['/', ',']:
1857 tokens
.restore_last_token()
1859 elif inside_choice
and string
== ',':
1860 tokens
.restore_last_token()
1863 if not current_selector
:
1864 raise syntax_error('"," must follow a format selector', start
)
1865 selectors
.append(current_selector
)
1866 current_selector
= None
1868 if not current_selector
:
1869 raise syntax_error('"/" must follow a format selector', start
)
1870 first_choice
= current_selector
1871 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1872 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1874 if not current_selector
:
1875 current_selector
= FormatSelector(SINGLE
, 'best', [])
1876 format_filter
= _parse_filter(tokens
)
1877 current_selector
.filters
.append(format_filter
)
1879 if current_selector
:
1880 raise syntax_error('Unexpected "("', start
)
1881 group
= _parse_format_selection(tokens
, inside_group
=True)
1882 current_selector
= FormatSelector(GROUP
, group
, [])
1884 if not current_selector
:
1885 raise syntax_error('Unexpected "+"', start
)
1886 selector_1
= current_selector
1887 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1889 raise syntax_error('Expected a selector', start
)
1890 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
1892 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
1893 elif type == tokenize
.ENDMARKER
:
1895 if current_selector
:
1896 selectors
.append(current_selector
)
1899 def _merge(formats_pair
):
1900 format_1
, format_2
= formats_pair
1903 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
1904 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
1906 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
1907 get_no_more
= {'video': False, 'audio': False}
1908 for (i
, fmt_info
) in enumerate(formats_info
):
1909 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
1912 for aud_vid
in ['audio', 'video']:
1913 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
1914 if get_no_more
[aud_vid
]:
1917 get_no_more
[aud_vid
] = True
1919 if len(formats_info
) == 1:
1920 return formats_info
[0]
1922 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
1923 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
1925 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
1926 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
1928 output_ext
= self
.params
.get('merge_output_format')
1931 output_ext
= the_only_video
['ext']
1932 elif the_only_audio
and not video_fmts
:
1933 output_ext
= the_only_audio
['ext']
1937 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
1940 'requested_formats': formats_info
,
1941 'format': '+'.join(filtered('format')),
1942 'format_id': '+'.join(filtered('format_id')),
1944 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
1945 'language': '+'.join(orderedSet(filtered('language'))) or None,
1946 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
1947 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
1948 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
1953 'width': the_only_video
.get('width'),
1954 'height': the_only_video
.get('height'),
1955 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
1956 'fps': the_only_video
.get('fps'),
1957 'dynamic_range': the_only_video
.get('dynamic_range'),
1958 'vcodec': the_only_video
.get('vcodec'),
1959 'vbr': the_only_video
.get('vbr'),
1960 'stretched_ratio': the_only_video
.get('stretched_ratio'),
1965 'acodec': the_only_audio
.get('acodec'),
1966 'abr': the_only_audio
.get('abr'),
1967 'asr': the_only_audio
.get('asr'),
1972 def _check_formats(formats
):
1973 if not check_formats
:
1976 yield from self
._check
_formats
(formats
)
1978 def _build_selector_function(selector
):
1979 if isinstance(selector
, list): # ,
1980 fs
= [_build_selector_function(s
) for s
in selector
]
1982 def selector_function(ctx
):
1985 return selector_function
1987 elif selector
.type == GROUP
: # ()
1988 selector_function
= _build_selector_function(selector
.selector
)
1990 elif selector
.type == PICKFIRST
: # /
1991 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
1993 def selector_function(ctx
):
1995 picked_formats
= list(f(ctx
))
1997 return picked_formats
2000 elif selector
.type == MERGE
: # +
2001 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2003 def selector_function(ctx
):
2004 for pair
in itertools
.product(
2005 selector_1(copy
.deepcopy(ctx
)), selector_2(copy
.deepcopy(ctx
))):
2008 elif selector
.type == SINGLE
: # atom
2009 format_spec
= selector
.selector
or 'best'
2011 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2012 if format_spec
== 'all':
2013 def selector_function(ctx
):
2014 yield from _check_formats(ctx
['formats'][::-1])
2015 elif format_spec
== 'mergeall':
2016 def selector_function(ctx
):
2017 formats
= list(_check_formats(ctx
['formats']))
2020 merged_format
= formats
[-1]
2021 for f
in formats
[-2::-1]:
2022 merged_format
= _merge((merged_format
, f
))
2026 format_fallback
, format_reverse
, format_idx
= False, True, 1
2028 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2030 if mobj
is not None:
2031 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2032 format_reverse
= mobj
.group('bw')[0] == 'b'
2033 format_type
= (mobj
.group('type') or [None])[0]
2034 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2035 format_modified
= mobj
.group('mod') is not None
2037 format_fallback
= not format_type
and not format_modified
# for b, w
2039 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2040 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2041 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2042 if format_type
# bv, ba, wv, wa
2043 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2044 if not format_modified
# b, w
2045 else lambda f
: True) # b*, w*
2046 filter_f
= lambda f
: _filter_f(f
) and (
2047 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2049 if format_spec
in self
._format
_selection
_exts
['audio']:
2050 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2051 elif format_spec
in self
._format
_selection
_exts
['video']:
2052 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2053 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2054 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2056 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2058 def selector_function(ctx
):
2059 formats
= list(ctx
['formats'])
2060 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2061 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
2062 # for extractors with incomplete formats (audio only (soundcloud)
2063 # or video only (imgur)) best/worst will fallback to
2064 # best/worst {video,audio}-only format
2066 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2068 yield matches
[format_idx
- 1]
2072 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2074 def final_selector(ctx
):
2075 ctx_copy
= copy
.deepcopy(ctx
)
2076 for _filter
in filters
:
2077 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2078 return selector_function(ctx_copy
)
2079 return final_selector
2081 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
2083 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
2084 except tokenize
.TokenError
:
2085 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2087 class TokenIterator(object):
2088 def __init__(self
, tokens
):
2089 self
.tokens
= tokens
2096 if self
.counter
>= len(self
.tokens
):
2097 raise StopIteration()
2098 value
= self
.tokens
[self
.counter
]
2104 def restore_last_token(self
):
2107 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2108 return _build_selector_function(parsed_selector
)
2110 def _calc_headers(self
, info_dict
):
2111 res
= std_headers
.copy()
2113 add_headers
= info_dict
.get('http_headers')
2115 res
.update(add_headers
)
2117 cookies
= self
._calc
_cookies
(info_dict
)
2119 res
['Cookie'] = cookies
2121 if 'X-Forwarded-For' not in res
:
2122 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2123 if x_forwarded_for_ip
:
2124 res
['X-Forwarded-For'] = x_forwarded_for_ip
2128 def _calc_cookies(self
, info_dict
):
2129 pr
= sanitized_Request(info_dict
['url'])
2130 self
.cookiejar
.add_cookie_header(pr
)
2131 return pr
.get_header('Cookie')
2133 def _sort_thumbnails(self
, thumbnails
):
2134 thumbnails
.sort(key
=lambda t
: (
2135 t
.get('preference') if t
.get('preference') is not None else -1,
2136 t
.get('width') if t
.get('width') is not None else -1,
2137 t
.get('height') if t
.get('height') is not None else -1,
2138 t
.get('id') if t
.get('id') is not None else '',
2141 def _sanitize_thumbnails(self
, info_dict
):
2142 thumbnails
= info_dict
.get('thumbnails')
2143 if thumbnails
is None:
2144 thumbnail
= info_dict
.get('thumbnail')
2146 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2150 def check_thumbnails(thumbnails
):
2151 for t
in thumbnails
:
2152 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2154 self
.urlopen(HEADRequest(t
['url']))
2155 except network_exceptions
as err
:
2156 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2160 self
._sort
_thumbnails
(thumbnails
)
2161 for i
, t
in enumerate(thumbnails
):
2162 if t
.get('id') is None:
2164 if t
.get('width') and t
.get('height'):
2165 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2166 t
['url'] = sanitize_url(t
['url'])
2168 if self
.params
.get('check_formats') is True:
2169 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1])).reverse()
2171 info_dict
['thumbnails'] = thumbnails
2173 def process_video_result(self
, info_dict
, download
=True):
2174 assert info_dict
.get('_type', 'video') == 'video'
2176 if 'id' not in info_dict
:
2177 raise ExtractorError('Missing "id" field in extractor result')
2178 if 'title' not in info_dict
:
2179 raise ExtractorError('Missing "title" field in extractor result',
2180 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2182 def report_force_conversion(field
, field_not
, conversion
):
2183 self
.report_warning(
2184 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2185 % (field
, field_not
, conversion
))
2187 def sanitize_string_field(info
, string_field
):
2188 field
= info
.get(string_field
)
2189 if field
is None or isinstance(field
, compat_str
):
2191 report_force_conversion(string_field
, 'a string', 'string')
2192 info
[string_field
] = compat_str(field
)
2194 def sanitize_numeric_fields(info
):
2195 for numeric_field
in self
._NUMERIC
_FIELDS
:
2196 field
= info
.get(numeric_field
)
2197 if field
is None or isinstance(field
, compat_numeric_types
):
2199 report_force_conversion(numeric_field
, 'numeric', 'int')
2200 info
[numeric_field
] = int_or_none(field
)
2202 sanitize_string_field(info_dict
, 'id')
2203 sanitize_numeric_fields(info_dict
)
2205 if 'playlist' not in info_dict
:
2206 # It isn't part of a playlist
2207 info_dict
['playlist'] = None
2208 info_dict
['playlist_index'] = None
2210 self
._sanitize
_thumbnails
(info_dict
)
2212 thumbnail
= info_dict
.get('thumbnail')
2213 thumbnails
= info_dict
.get('thumbnails')
2215 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2217 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2219 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2220 info_dict
['display_id'] = info_dict
['id']
2222 if info_dict
.get('duration') is not None:
2223 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2225 for ts_key
, date_key
in (
2226 ('timestamp', 'upload_date'),
2227 ('release_timestamp', 'release_date'),
2229 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2230 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2231 # see http://bugs.python.org/issue1646728)
2233 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2234 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2235 except (ValueError, OverflowError, OSError):
2238 live_keys
= ('is_live', 'was_live')
2239 live_status
= info_dict
.get('live_status')
2240 if live_status
is None:
2241 for key
in live_keys
:
2242 if info_dict
.get(key
) is False:
2244 if info_dict
.get(key
):
2247 if all(info_dict
.get(key
) is False for key
in live_keys
):
2248 live_status
= 'not_live'
2250 info_dict
['live_status'] = live_status
2251 for key
in live_keys
:
2252 if info_dict
.get(key
) is None:
2253 info_dict
[key
] = (live_status
== key
)
2255 # Auto generate title fields corresponding to the *_number fields when missing
2256 # in order to always have clean titles. This is very common for TV series.
2257 for field
in ('chapter', 'season', 'episode'):
2258 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2259 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2261 for cc_kind
in ('subtitles', 'automatic_captions'):
2262 cc
= info_dict
.get(cc_kind
)
2264 for _
, subtitle
in cc
.items():
2265 for subtitle_format
in subtitle
:
2266 if subtitle_format
.get('url'):
2267 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2268 if subtitle_format
.get('ext') is None:
2269 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2271 automatic_captions
= info_dict
.get('automatic_captions')
2272 subtitles
= info_dict
.get('subtitles')
2274 info_dict
['requested_subtitles'] = self
.process_subtitles(
2275 info_dict
['id'], subtitles
, automatic_captions
)
2277 if info_dict
.get('formats') is None:
2278 # There's only one format available
2279 formats
= [info_dict
]
2281 formats
= info_dict
['formats']
2283 info_dict
['__has_drm'] = any(f
.get('has_drm') for f
in formats
)
2284 if not self
.params
.get('allow_unplayable_formats'):
2285 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2288 self
.raise_no_formats(info_dict
)
2290 def is_wellformed(f
):
2293 self
.report_warning(
2294 '"url" field is missing or empty - skipping format, '
2295 'there is an error in extractor')
2297 if isinstance(url
, bytes):
2298 sanitize_string_field(f
, 'url')
2301 # Filter out malformed formats for better extraction robustness
2302 formats
= list(filter(is_wellformed
, formats
))
2306 # We check that all the formats have the format and format_id fields
2307 for i
, format
in enumerate(formats
):
2308 sanitize_string_field(format
, 'format_id')
2309 sanitize_numeric_fields(format
)
2310 format
['url'] = sanitize_url(format
['url'])
2311 if not format
.get('format_id'):
2312 format
['format_id'] = compat_str(i
)
2314 # Sanitize format_id from characters used in format selector expression
2315 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2316 format_id
= format
['format_id']
2317 if format_id
not in formats_dict
:
2318 formats_dict
[format_id
] = []
2319 formats_dict
[format_id
].append(format
)
2321 # Make sure all formats have unique format_id
2322 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2323 for format_id
, ambiguous_formats
in formats_dict
.items():
2324 ambigious_id
= len(ambiguous_formats
) > 1
2325 for i
, format
in enumerate(ambiguous_formats
):
2327 format
['format_id'] = '%s-%d' % (format_id
, i
)
2328 if format
.get('ext') is None:
2329 format
['ext'] = determine_ext(format
['url']).lower()
2330 # Ensure there is no conflict between id and ext in format selection
2331 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2332 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2333 format
['format_id'] = 'f%s' % format
['format_id']
2335 for i
, format
in enumerate(formats
):
2336 if format
.get('format') is None:
2337 format
['format'] = '{id} - {res}{note}'.format(
2338 id=format
['format_id'],
2339 res
=self
.format_resolution(format
),
2340 note
=format_field(format
, 'format_note', ' (%s)'),
2342 if format
.get('protocol') is None:
2343 format
['protocol'] = determine_protocol(format
)
2344 if format
.get('resolution') is None:
2345 format
['resolution'] = self
.format_resolution(format
, default
=None)
2346 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2347 format
['dynamic_range'] = 'SDR'
2348 if (info_dict
.get('duration') and format
.get('tbr')
2349 and not format
.get('filesize') and not format
.get('filesize_approx')):
2350 format
['filesize_approx'] = info_dict
['duration'] * format
['tbr'] * (1024 / 8)
2352 # Add HTTP headers, so that external programs can use them from the
2354 full_format_info
= info_dict
.copy()
2355 full_format_info
.update(format
)
2356 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2357 # Remove private housekeeping stuff
2358 if '__x_forwarded_for_ip' in info_dict
:
2359 del info_dict
['__x_forwarded_for_ip']
2361 # TODO Central sorting goes here
2363 if self
.params
.get('check_formats') is True:
2364 formats
= LazyList(self
._check
_formats
(formats
[::-1])).reverse()
2366 if not formats
or formats
[0] is not info_dict
:
2367 # only set the 'formats' fields if the original info_dict list them
2368 # otherwise we end up with a circular reference, the first (and unique)
2369 # element in the 'formats' field in info_dict is info_dict itself,
2370 # which can't be exported to json
2371 info_dict
['formats'] = formats
2373 info_dict
, _
= self
.pre_process(info_dict
)
2375 # The pre-processors may have modified the formats
2376 formats
= info_dict
.get('formats', [info_dict
])
2378 if self
.params
.get('list_thumbnails'):
2379 self
.list_thumbnails(info_dict
)
2380 if self
.params
.get('listformats'):
2381 if not info_dict
.get('formats') and not info_dict
.get('url'):
2382 self
.to_screen('%s has no formats' % info_dict
['id'])
2384 self
.list_formats(info_dict
)
2385 if self
.params
.get('listsubtitles'):
2386 if 'automatic_captions' in info_dict
:
2387 self
.list_subtitles(
2388 info_dict
['id'], automatic_captions
, 'automatic captions')
2389 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2390 list_only
= self
.params
.get('simulate') is None and (
2391 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2393 # Without this printing, -F --print-json will not work
2394 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2397 format_selector
= self
.format_selector
2398 if format_selector
is None:
2399 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2400 self
.write_debug('Default format spec: %s' % req_format
)
2401 format_selector
= self
.build_format_selector(req_format
)
2403 # While in format selection we may need to have an access to the original
2404 # format set in order to calculate some metrics or do some processing.
2405 # For now we need to be able to guess whether original formats provided
2406 # by extractor are incomplete or not (i.e. whether extractor provides only
2407 # video-only or audio-only formats) for proper formats selection for
2408 # extractors with such incomplete formats (see
2409 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2410 # Since formats may be filtered during format selection and may not match
2411 # the original formats the results may be incorrect. Thus original formats
2412 # or pre-calculated metrics should be passed to format selection routines
2414 # We will pass a context object containing all necessary additional data
2415 # instead of just formats.
2416 # This fixes incorrect format selection issue (see
2417 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2418 incomplete_formats
= (
2419 # All formats are video-only or
2420 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2421 # all formats are audio-only
2422 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2426 'incomplete_formats': incomplete_formats
,
2429 formats_to_download
= list(format_selector(ctx
))
2430 if not formats_to_download
:
2431 if not self
.params
.get('ignore_no_formats_error'):
2432 raise ExtractorError('Requested format is not available', expected
=True,
2433 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2435 self
.report_warning('Requested format is not available')
2436 # Process what we can, even without any available formats.
2437 self
.process_info(dict(info_dict
))
2440 '[info] %s: Downloading %d format(s): %s' % (
2441 info_dict
['id'], len(formats_to_download
),
2442 ", ".join([f
['format_id'] for f
in formats_to_download
])))
2443 for fmt
in formats_to_download
:
2444 new_info
= dict(info_dict
)
2445 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2446 new_info
['__original_infodict'] = info_dict
2447 new_info
.update(fmt
)
2448 self
.process_info(new_info
)
2449 # We update the info dict with the selected best quality format (backwards compatibility)
2450 if formats_to_download
:
2451 info_dict
.update(formats_to_download
[-1])
2454 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2455 """Select the requested subtitles and their format"""
2457 if normal_subtitles
and self
.params
.get('writesubtitles'):
2458 available_subs
.update(normal_subtitles
)
2459 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2460 for lang
, cap_info
in automatic_captions
.items():
2461 if lang
not in available_subs
:
2462 available_subs
[lang
] = cap_info
2464 if (not self
.params
.get('writesubtitles') and not
2465 self
.params
.get('writeautomaticsub') or not
2469 all_sub_langs
= available_subs
.keys()
2470 if self
.params
.get('allsubtitles', False):
2471 requested_langs
= all_sub_langs
2472 elif self
.params
.get('subtitleslangs', False):
2473 # A list is used so that the order of languages will be the same as
2474 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2475 requested_langs
= []
2476 for lang_re
in self
.params
.get('subtitleslangs'):
2477 if lang_re
== 'all':
2478 requested_langs
.extend(all_sub_langs
)
2480 discard
= lang_re
[0] == '-'
2482 lang_re
= lang_re
[1:]
2483 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2485 for lang
in current_langs
:
2486 while lang
in requested_langs
:
2487 requested_langs
.remove(lang
)
2489 requested_langs
.extend(current_langs
)
2490 requested_langs
= orderedSet(requested_langs
)
2491 elif 'en' in available_subs
:
2492 requested_langs
= ['en']
2494 requested_langs
= [list(all_sub_langs
)[0]]
2496 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2498 formats_query
= self
.params
.get('subtitlesformat', 'best')
2499 formats_preference
= formats_query
.split('/') if formats_query
else []
2501 for lang
in requested_langs
:
2502 formats
= available_subs
.get(lang
)
2504 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2506 for ext
in formats_preference
:
2510 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2516 self
.report_warning(
2517 'No subtitle format found matching "%s" for language %s, '
2518 'using %s' % (formats_query
, lang
, f
['ext']))
2522 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2523 def print_mandatory(field
, actual_field
=None):
2524 if actual_field
is None:
2525 actual_field
= field
2526 if (self
.params
.get('force%s' % field
, False)
2527 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2528 self
.to_stdout(info_dict
[actual_field
])
2530 def print_optional(field
):
2531 if (self
.params
.get('force%s' % field
, False)
2532 and info_dict
.get(field
) is not None):
2533 self
.to_stdout(info_dict
[field
])
2535 info_dict
= info_dict
.copy()
2536 if filename
is not None:
2537 info_dict
['filename'] = filename
2538 if info_dict
.get('requested_formats') is not None:
2539 # For RTMP URLs, also include the playpath
2540 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2541 elif 'url' in info_dict
:
2542 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2544 if self
.params
.get('forceprint') or self
.params
.get('forcejson'):
2545 self
.post_extract(info_dict
)
2546 for tmpl
in self
.params
.get('forceprint', []):
2547 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2548 if mobj
and mobj
.group(1):
2549 tmpl
= f
'{tmpl[:-1]} = %({tmpl[:-1]})s'
2551 tmpl
= '%({})s'.format(tmpl
)
2552 self
.to_stdout(self
.evaluate_outtmpl(tmpl
, info_dict
))
2554 print_mandatory('title')
2555 print_mandatory('id')
2556 print_mandatory('url', 'urls')
2557 print_optional('thumbnail')
2558 print_optional('description')
2559 print_optional('filename')
2560 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2561 self
.to_stdout(formatSeconds(info_dict
['duration']))
2562 print_mandatory('format')
2564 if self
.params
.get('forcejson'):
2565 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2567 def dl(self
, name
, info
, subtitle
=False, test
=False):
2568 if not info
.get('url'):
2569 self
.raise_no_formats(info
, True)
2572 verbose
= self
.params
.get('verbose')
2575 'quiet': self
.params
.get('quiet') or not verbose
,
2577 'noprogress': not verbose
,
2579 'skip_unavailable_fragments': False,
2580 'keep_fragments': False,
2582 '_no_ytdl_file': True,
2585 params
= self
.params
2586 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2588 for ph
in self
._progress
_hooks
:
2589 fd
.add_progress_hook(ph
)
2590 urls
= '", "'.join([f
['url'] for f
in info
.get('requested_formats', [])] or [info
['url']])
2591 self
.write_debug('Invoking downloader on "%s"' % urls
)
2593 new_info
= copy
.deepcopy(self
._copy
_infodict
(info
))
2594 if new_info
.get('http_headers') is None:
2595 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2596 return fd
.download(name
, new_info
, subtitle
)
2598 def process_info(self
, info_dict
):
2599 """Process a single resolved IE result."""
2601 assert info_dict
.get('_type', 'video') == 'video'
2603 max_downloads
= self
.params
.get('max_downloads')
2604 if max_downloads
is not None:
2605 if self
._num
_downloads
>= int(max_downloads
):
2606 raise MaxDownloadsReached()
2608 # TODO: backward compatibility, to be removed
2609 info_dict
['fulltitle'] = info_dict
['title']
2611 if 'format' not in info_dict
and 'ext' in info_dict
:
2612 info_dict
['format'] = info_dict
['ext']
2614 if self
._match
_entry
(info_dict
) is not None:
2617 self
.post_extract(info_dict
)
2618 self
._num
_downloads
+= 1
2620 # info_dict['_filename'] needs to be set for backward compatibility
2621 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2622 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2626 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2628 if self
.params
.get('simulate'):
2629 if self
.params
.get('force_write_download_archive', False):
2630 self
.record_download_archive(info_dict
)
2631 # Do nothing else if in simulate mode
2634 if full_filename
is None:
2636 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2638 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2641 if self
._write
_description
('video', info_dict
,
2642 self
.prepare_filename(info_dict
, 'description')) is None:
2645 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2646 if sub_files
is None:
2648 files_to_move
.update(dict(sub_files
))
2650 thumb_files
= self
._write
_thumbnails
(
2651 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2652 if thumb_files
is None:
2654 files_to_move
.update(dict(thumb_files
))
2656 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2657 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2658 if _infojson_written
:
2659 info_dict
['infojson_filename'] = infofn
2660 # For backward compatability, even though it was a private field
2661 info_dict
['__infojson_filename'] = infofn
2662 elif _infojson_written
is None:
2665 # Note: Annotations are deprecated
2667 if self
.params
.get('writeannotations', False):
2668 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2670 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2672 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2673 self
.to_screen('[info] Video annotations are already present')
2674 elif not info_dict
.get('annotations'):
2675 self
.report_warning('There are no annotations to write.')
2678 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2679 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2680 annofile
.write(info_dict
['annotations'])
2681 except (KeyError, TypeError):
2682 self
.report_warning('There are no annotations to write.')
2683 except (OSError, IOError):
2684 self
.report_error('Cannot write annotations file: ' + annofn
)
2687 # Write internet shortcut files
2688 def _write_link_file(link_type
):
2689 if 'webpage_url' not in info_dict
:
2690 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2692 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2693 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2694 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2697 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2698 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2699 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2700 template_vars
= {'url': iri_to_uri(info_dict['webpage_url'])}
2701 if link_type
== 'desktop':
2702 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2703 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
2704 except (OSError, IOError):
2705 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
2710 'url': self
.params
.get('writeurllink'),
2711 'webloc': self
.params
.get('writewebloclink'),
2712 'desktop': self
.params
.get('writedesktoplink'),
2714 if self
.params
.get('writelink'):
2715 link_type
= ('webloc' if sys
.platform
== 'darwin'
2716 else 'desktop' if sys
.platform
.startswith('linux')
2718 write_links
[link_type
] = True
2720 if any(should_write
and not _write_link_file(link_type
)
2721 for link_type
, should_write
in write_links
.items()):
2725 info_dict
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2726 except PostProcessingError
as err
:
2727 self
.report_error('Preprocessing: %s' % str(err
))
2730 must_record_download_archive
= False
2731 if self
.params
.get('skip_download', False):
2732 info_dict
['filepath'] = temp_filename
2733 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2734 info_dict
['__files_to_move'] = files_to_move
2735 info_dict
= self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
)
2738 info_dict
.setdefault('__postprocessors', [])
2741 def existing_file(*filepaths
):
2742 ext
= info_dict
.get('ext')
2743 final_ext
= self
.params
.get('final_ext', ext
)
2745 for file in orderedSet(filepaths
):
2746 if final_ext
!= ext
:
2747 converted
= replace_extension(file, final_ext
, ext
)
2748 if os
.path
.exists(encodeFilename(converted
)):
2749 existing_files
.append(converted
)
2750 if os
.path
.exists(encodeFilename(file)):
2751 existing_files
.append(file)
2753 if not existing_files
or self
.params
.get('overwrites', False):
2754 for file in orderedSet(existing_files
):
2755 self
.report_file_delete(file)
2756 os
.remove(encodeFilename(file))
2759 info_dict
['ext'] = os
.path
.splitext(existing_files
[0])[1][1:]
2760 return existing_files
[0]
2763 if info_dict
.get('requested_formats') is not None:
2765 def compatible_formats(formats
):
2766 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2767 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2768 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2769 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2773 exts
= set(format
.get('ext') for format
in formats
)
2775 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2778 for ext_sets
in COMPATIBLE_EXTS
:
2779 if ext_sets
.issuperset(exts
):
2781 # TODO: Check acodec/vcodec
2784 requested_formats
= info_dict
['requested_formats']
2785 old_ext
= info_dict
['ext']
2786 if self
.params
.get('merge_output_format') is None:
2787 if not compatible_formats(requested_formats
):
2788 info_dict
['ext'] = 'mkv'
2789 self
.report_warning(
2790 'Requested formats are incompatible for merge and will be merged into mkv')
2791 if (info_dict
['ext'] == 'webm'
2792 and info_dict
.get('thumbnails')
2793 # check with type instead of pp_key, __name__, or isinstance
2794 # since we dont want any custom PPs to trigger this
2795 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])):
2796 info_dict
['ext'] = 'mkv'
2797 self
.report_warning(
2798 'webm doesn\'t support embedding a thumbnail, mkv will be used')
2799 new_ext
= info_dict
['ext']
2801 def correct_ext(filename
, ext
=new_ext
):
2804 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2806 os
.path
.splitext(filename
)[0]
2807 if filename_real_ext
in (old_ext
, new_ext
)
2809 return '%s.%s' % (filename_wo_ext
, ext
)
2811 # Ensure filename always has a correct extension for successful merge
2812 full_filename
= correct_ext(full_filename
)
2813 temp_filename
= correct_ext(temp_filename
)
2814 dl_filename
= existing_file(full_filename
, temp_filename
)
2815 info_dict
['__real_download'] = False
2817 if dl_filename
is not None:
2818 self
.report_file_already_downloaded(dl_filename
)
2819 elif get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-'):
2820 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
2821 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2822 info_dict
['__real_download'] = real_download
2825 merger
= FFmpegMergerPP(self
)
2826 if self
.params
.get('allow_unplayable_formats'):
2827 self
.report_warning(
2828 'You have requested merging of multiple formats '
2829 'while also allowing unplayable formats to be downloaded. '
2830 'The formats won\'t be merged to prevent data corruption.')
2831 elif not merger
.available
:
2832 self
.report_warning(
2833 'You have requested merging of multiple formats but ffmpeg is not installed. '
2834 'The formats won\'t be merged.')
2836 if temp_filename
== '-':
2837 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
)
2838 else 'but the formats are incompatible for simultaneous download' if merger
.available
2839 else 'but ffmpeg is not installed')
2840 self
.report_warning(
2841 f
'You have requested downloading multiple formats to stdout {reason}. '
2842 'The formats will be streamed one after the other')
2843 fname
= temp_filename
2844 for f
in requested_formats
:
2845 new_info
= dict(info_dict
)
2846 del new_info
['requested_formats']
2848 if temp_filename
!= '-':
2849 fname
= prepend_extension(
2850 correct_ext(temp_filename
, new_info
['ext']),
2851 'f%s' % f
['format_id'], new_info
['ext'])
2852 if not self
._ensure
_dir
_exists
(fname
):
2854 f
['filepath'] = fname
2855 downloaded
.append(fname
)
2856 partial_success
, real_download
= self
.dl(fname
, new_info
)
2857 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
2858 success
= success
and partial_success
2859 if merger
.available
and not self
.params
.get('allow_unplayable_formats'):
2860 info_dict
['__postprocessors'].append(merger
)
2861 info_dict
['__files_to_merge'] = downloaded
2862 # Even if there were no downloads, it is being merged only now
2863 info_dict
['__real_download'] = True
2865 for file in downloaded
:
2866 files_to_move
[file] = None
2868 # Just a single file
2869 dl_filename
= existing_file(full_filename
, temp_filename
)
2870 if dl_filename
is None or dl_filename
== temp_filename
:
2871 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
2872 # So we should try to resume the download
2873 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2874 info_dict
['__real_download'] = real_download
2876 self
.report_file_already_downloaded(dl_filename
)
2878 dl_filename
= dl_filename
or temp_filename
2879 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2881 except network_exceptions
as err
:
2882 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
2884 except (OSError, IOError) as err
:
2885 raise UnavailableVideoError(err
)
2886 except (ContentTooShortError
, ) as err
:
2887 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
2890 if success
and full_filename
!= '-':
2894 fixup_policy
= self
.params
.get('fixup')
2895 vid
= info_dict
['id']
2897 if fixup_policy
in ('ignore', 'never'):
2899 elif fixup_policy
== 'warn':
2901 elif fixup_policy
!= 'force':
2902 assert fixup_policy
in ('detect_or_warn', None)
2903 if not info_dict
.get('__real_download'):
2906 def ffmpeg_fixup(cndn
, msg
, cls
):
2910 self
.report_warning(f
'{vid}: {msg}')
2914 info_dict
['__postprocessors'].append(pp
)
2916 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
2918 stretched_ratio
= info_dict
.get('stretched_ratio')
2920 stretched_ratio
not in (1, None),
2921 f
'Non-uniform pixel ratio {stretched_ratio}',
2922 FFmpegFixupStretchedPP
)
2925 (info_dict
.get('requested_formats') is None
2926 and info_dict
.get('container') == 'm4a_dash'
2927 and info_dict
.get('ext') == 'm4a'),
2928 'writing DASH m4a. Only some players support this container',
2931 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
2932 downloader
= downloader
.__name
__ if downloader
else None
2933 ffmpeg_fixup(info_dict
.get('requested_formats') is None and downloader
== 'HlsFD',
2934 'malformed AAC bitstream detected', FFmpegFixupM3u8PP
)
2935 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP
)
2936 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP
)
2940 info_dict
= self
.post_process(dl_filename
, info_dict
, files_to_move
)
2941 except PostProcessingError
as err
:
2942 self
.report_error('Postprocessing: %s' % str(err
))
2945 for ph
in self
._post
_hooks
:
2946 ph(info_dict
['filepath'])
2947 except Exception as err
:
2948 self
.report_error('post hooks: %s' % str(err
))
2950 must_record_download_archive
= True
2952 if must_record_download_archive
or self
.params
.get('force_write_download_archive', False):
2953 self
.record_download_archive(info_dict
)
2954 max_downloads
= self
.params
.get('max_downloads')
2955 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
2956 raise MaxDownloadsReached()
2958 def __download_wrapper(self
, func
):
2959 @functools.wraps(func
)
2960 def wrapper(*args
, **kwargs
):
2962 res
= func(*args
, **kwargs
)
2963 except UnavailableVideoError
as e
:
2964 self
.report_error(e
)
2965 except DownloadCancelled
as e
:
2966 self
.to_screen(f
'[info] {e}')
2969 if self
.params
.get('dump_single_json', False):
2970 self
.post_extract(res
)
2971 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
2974 def download(self
, url_list
):
2975 """Download a given list of URLs."""
2976 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
2977 outtmpl
= self
.outtmpl_dict
['default']
2978 if (len(url_list
) > 1
2980 and '%' not in outtmpl
2981 and self
.params
.get('max_downloads') != 1):
2982 raise SameFileError(outtmpl
)
2984 for url
in url_list
:
2985 self
.__download
_wrapper
(self
.extract_info
)(
2986 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
2988 return self
._download
_retcode
2990 def download_with_info_file(self
, info_filename
):
2991 with contextlib
.closing(fileinput
.FileInput(
2992 [info_filename
], mode
='r',
2993 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
2994 # FileInput doesn't have a read method, we can't call json.load
2995 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
2997 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
2998 except (DownloadError
, EntryNotInPlaylist
, ThrottledDownload
) as e
:
2999 if not isinstance(e
, EntryNotInPlaylist
):
3000 self
.to_stderr('\r')
3001 webpage_url
= info
.get('webpage_url')
3002 if webpage_url
is not None:
3003 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3004 return self
.download([webpage_url
])
3007 return self
._download
_retcode
3010 def sanitize_info(info_dict
, remove_private_keys
=False):
3011 ''' Sanitize the infodict for converting to json '''
3012 if info_dict
is None:
3014 info_dict
.setdefault('epoch', int(time
.time()))
3015 remove_keys
= {'__original_infodict'}
# Always remove this since this may contain a copy of the entire dict
3016 keep_keys
= ['_type'] # Always keep this to facilitate load-info-json
3017 if remove_private_keys
:
3019 'requested_formats', 'requested_subtitles', 'requested_entries', 'entries',
3020 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3022 empty_values
= (None, {}, [], set(), tuple())
3023 reject
= lambda k
, v
: k
not in keep_keys
and (
3024 k
.startswith('_') or k
in remove_keys
or v
in empty_values
)
3026 reject
= lambda k
, v
: k
in remove_keys
3027 filter_fn
= lambda obj
: (
3028 list(map(filter_fn
, obj
)) if isinstance(obj
, (LazyList
, list, tuple, set))
3029 else obj
if not isinstance(obj
, dict)
3030 else dict((k
, filter_fn(v
)) for k
, v
in obj
.items() if not reject(k
, v
)))
3031 return filter_fn(info_dict
)
3034 def filter_requested_info(info_dict
, actually_filter
=True):
3035 ''' Alias of sanitize_info for backward compatibility '''
3036 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3038 def run_pp(self
, pp
, infodict
):
3039 files_to_delete
= []
3040 if '__files_to_move' not in infodict
:
3041 infodict
['__files_to_move'] = {}
3043 files_to_delete
, infodict
= pp
.run(infodict
)
3044 except PostProcessingError
as e
:
3045 # Must be True and not 'only_download'
3046 if self
.params
.get('ignoreerrors') is True:
3047 self
.report_error(e
)
3051 if not files_to_delete
:
3053 if self
.params
.get('keepvideo', False):
3054 for f
in files_to_delete
:
3055 infodict
['__files_to_move'].setdefault(f
, '')
3057 for old_filename
in set(files_to_delete
):
3058 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
3060 os
.remove(encodeFilename(old_filename
))
3061 except (IOError, OSError):
3062 self
.report_warning('Unable to remove downloaded original file')
3063 if old_filename
in infodict
['__files_to_move']:
3064 del infodict
['__files_to_move'][old_filename
]
3068 def post_extract(info_dict
):
3069 def actual_post_extract(info_dict
):
3070 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3071 for video_dict
in info_dict
.get('entries', {}):
3072 actual_post_extract(video_dict
or {})
3075 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
3076 extra
= post_extractor().items()
3077 info_dict
.update(extra
)
3078 info_dict
.pop('__post_extractor', None)
3080 original_infodict
= info_dict
.get('__original_infodict') or {}
3081 original_infodict
.update(extra
)
3082 original_infodict
.pop('__post_extractor', None)
3084 actual_post_extract(info_dict
or {})
3086 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3087 info
= dict(ie_info
)
3088 info
['__files_to_move'] = files_to_move
or {}
3089 for pp
in self
._pps
[key
]:
3090 info
= self
.run_pp(pp
, info
)
3091 return info
, info
.pop('__files_to_move', None)
3093 def post_process(self
, filename
, ie_info
, files_to_move
=None):
3094 """Run all the postprocessors on the given file."""
3095 info
= dict(ie_info
)
3096 info
['filepath'] = filename
3097 info
['__files_to_move'] = files_to_move
or {}
3099 for pp
in ie_info
.get('__postprocessors', []) + self
._pps
['post_process']:
3100 info
= self
.run_pp(pp
, info
)
3101 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3102 del info
['__files_to_move']
3103 for pp
in self
._pps
['after_move']:
3104 info
= self
.run_pp(pp
, info
)
3107 def _make_archive_id(self
, info_dict
):
3108 video_id
= info_dict
.get('id')
3111 # Future-proof against any change in case
3112 # and backwards compatibility with prior versions
3113 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3114 if extractor
is None:
3115 url
= str_or_none(info_dict
.get('url'))
3118 # Try to find matching extractor for the URL and take its ie_key
3119 for ie_key
, ie
in self
._ies
.items():
3120 if ie
.suitable(url
):
3125 return '%s %s' % (extractor
.lower(), video_id
)
3127 def in_download_archive(self
, info_dict
):
3128 fn
= self
.params
.get('download_archive')
3132 vid_id
= self
._make
_archive
_id
(info_dict
)
3134 return False # Incomplete video information
3136 return vid_id
in self
.archive
3138 def record_download_archive(self
, info_dict
):
3139 fn
= self
.params
.get('download_archive')
3142 vid_id
= self
._make
_archive
_id
(info_dict
)
3144 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3145 archive_file
.write(vid_id
+ '\n')
3146 self
.archive
.add(vid_id
)
3149 def format_resolution(format
, default
='unknown'):
3150 is_images
= format
.get('vcodec') == 'none' and format
.get('acodec') == 'none'
3151 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3153 if format
.get('resolution') is not None:
3154 return format
['resolution']
3155 if format
.get('width') and format
.get('height'):
3156 res
= '%dx%d' % (format
['width'], format
['height'])
3157 elif format
.get('height'):
3158 res
= '%sp' % format
['height']
3159 elif format
.get('width'):
3160 res
= '%dx?' % format
['width']
3165 return f
'img {res}' if is_images
else res
3167 def _format_note(self
, fdict
):
3169 if fdict
.get('ext') in ['f4f', 'f4m']:
3170 res
+= '(unsupported) '
3171 if fdict
.get('language'):
3174 res
+= '[%s] ' % fdict
['language']
3175 if fdict
.get('format_note') is not None:
3176 res
+= fdict
['format_note'] + ' '
3177 if fdict
.get('tbr') is not None:
3178 res
+= '%4dk ' % fdict
['tbr']
3179 if fdict
.get('container') is not None:
3182 res
+= '%s container' % fdict
['container']
3183 if (fdict
.get('vcodec') is not None
3184 and fdict
.get('vcodec') != 'none'):
3187 res
+= fdict
['vcodec']
3188 if fdict
.get('vbr') is not None:
3190 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3192 if fdict
.get('vbr') is not None:
3193 res
+= '%4dk' % fdict
['vbr']
3194 if fdict
.get('fps') is not None:
3197 res
+= '%sfps' % fdict
['fps']
3198 if fdict
.get('acodec') is not None:
3201 if fdict
['acodec'] == 'none':
3204 res
+= '%-5s' % fdict
['acodec']
3205 elif fdict
.get('abr') is not None:
3209 if fdict
.get('abr') is not None:
3210 res
+= '@%3dk' % fdict
['abr']
3211 if fdict
.get('asr') is not None:
3212 res
+= ' (%5dHz)' % fdict
['asr']
3213 if fdict
.get('filesize') is not None:
3216 res
+= format_bytes(fdict
['filesize'])
3217 elif fdict
.get('filesize_approx') is not None:
3220 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3223 def _list_format_headers(self
, *headers
):
3224 if self
.params
.get('listformats_table', True) is not False:
3225 return [self
._format
_screen
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3228 def list_formats(self
, info_dict
):
3229 formats
= info_dict
.get('formats', [info_dict
])
3230 new_format
= self
.params
.get('listformats_table', True) is not False
3232 tbr_digits
= number_of_digits(max(f
.get('tbr') or 0 for f
in formats
))
3233 vbr_digits
= number_of_digits(max(f
.get('vbr') or 0 for f
in formats
))
3234 abr_digits
= number_of_digits(max(f
.get('abr') or 0 for f
in formats
))
3235 delim
= self
._format
_screen
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3238 self
._format
_screen
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3239 format_field(f
, 'ext'),
3240 self
.format_resolution(f
),
3241 format_field(f
, 'fps', '%3d'),
3242 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3244 format_field(f
, 'filesize', ' %s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~%s', func
=format_bytes
),
3245 format_field(f
, 'tbr', f
'%{tbr_digits}dk'),
3246 shorten_protocol_name(f
.get('protocol', '').replace("native", "n")),
3248 format_field(f
, 'vcodec', default
='unknown').replace('none', ''),
3249 format_field(f
, 'vbr', f
'%{vbr_digits}dk'),
3250 format_field(f
, 'acodec', default
='unknown').replace('none', ''),
3251 format_field(f
, 'abr', f
'%{abr_digits}dk'),
3252 format_field(f
, 'asr', '%5dHz'),
3254 self
._format
_screen
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3255 format_field(f
, 'language', '[%s]'),
3256 format_field(f
, 'format_note'),
3257 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3259 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3260 header_line
= self
._list
_format
_headers
(
3261 'ID', 'EXT', 'RESOLUTION', 'FPS', 'HDR', delim
, ' FILESIZE', ' TBR', 'PROTO',
3262 delim
, 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO')
3266 format_field(f
, 'format_id'),
3267 format_field(f
, 'ext'),
3268 self
.format_resolution(f
),
3269 self
._format
_note
(f
)]
3271 if f
.get('preference') is None or f
['preference'] >= -1000]
3272 header_line
= ['format code', 'extension', 'resolution', 'note']
3275 '[info] Available formats for %s:' % info_dict
['id'])
3276 self
.to_stdout(render_table(
3278 extraGap
=(0 if new_format
else 1),
3279 hideEmpty
=new_format
,
3280 delim
=new_format
and self
._format
_screen
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True)))
3282 def list_thumbnails(self
, info_dict
):
3283 thumbnails
= list(info_dict
.get('thumbnails'))
3285 self
.to_screen('[info] No thumbnails present for %s' % info_dict
['id'])
3289 '[info] Thumbnails for %s:' % info_dict
['id'])
3290 self
.to_stdout(render_table(
3291 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3292 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
]))
3294 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3296 self
.to_screen('%s has no %s' % (video_id
, name
))
3299 'Available %s for %s:' % (name
, video_id
))
3301 def _row(lang
, formats
):
3302 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3303 if len(set(names
)) == 1:
3304 names
= [] if names
[0] == 'unknown' else names
[:1]
3305 return [lang
, ', '.join(names
), ', '.join(exts
)]
3307 self
.to_stdout(render_table(
3308 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3309 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3312 def urlopen(self
, req
):
3313 """ Start an HTTP download """
3314 if isinstance(req
, compat_basestring
):
3315 req
= sanitized_Request(req
)
3316 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3318 def print_debug_header(self
):
3319 if not self
.params
.get('verbose'):
3322 def get_encoding(stream
):
3323 ret
= getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__)
3324 if not supports_terminal_sequences(stream
):
3328 encoding_str
= 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3329 locale
.getpreferredencoding(),
3330 sys
.getfilesystemencoding(),
3331 get_encoding(self
._screen
_file
), get_encoding(self
._err
_file
),
3332 self
.get_encoding())
3334 logger
= self
.params
.get('logger')
3336 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3337 write_debug(encoding_str
)
3339 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3340 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3342 source
= detect_variant()
3343 write_debug('yt-dlp version %s%s' % (__version__
, '' if source
== 'unknown' else f
' ({source})'))
3344 if not _LAZY_LOADER
:
3345 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3346 write_debug('Lazy loading extractors is forcibly disabled')
3348 write_debug('Lazy loading extractors is disabled')
3349 if plugin_extractors
or plugin_postprocessors
:
3350 write_debug('Plugins: %s' % [
3351 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3352 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3353 if self
.params
.get('compat_opts'):
3354 write_debug('Compatibility options: %s' % ', '.join(self
.params
.get('compat_opts')))
3357 ['git', 'rev-parse', '--short', 'HEAD'],
3358 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3359 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3360 out
, err
= sp
.communicate_or_kill()
3361 out
= out
.decode().strip()
3362 if re
.match('[0-9a-f]+', out
):
3363 write_debug('Git HEAD: %s' % out
)
3370 def python_implementation():
3371 impl_name
= platform
.python_implementation()
3372 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3373 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3376 write_debug('Python version %s (%s %s) - %s' % (
3377 platform
.python_version(),
3378 python_implementation(),
3379 platform
.architecture()[0],
3382 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3383 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3385 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features
)
3387 exe_versions
['rtmpdump'] = rtmpdump_version()
3388 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3389 exe_str
= ', '.join(
3390 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3392 write_debug('exe versions: %s' % exe_str
)
3394 from .downloader
.websocket
import has_websockets
3395 from .postprocessor
.embedthumbnail
import has_mutagen
3396 from .cookies
import SQLITE_AVAILABLE
, KEYRING_AVAILABLE
3398 lib_str
= join_nonempty(
3399 compat_pycrypto_AES
and compat_pycrypto_AES
.__name
__.split('.')[0],
3400 KEYRING_AVAILABLE
and 'keyring',
3401 has_mutagen
and 'mutagen',
3402 SQLITE_AVAILABLE
and 'sqlite',
3403 has_websockets
and 'websockets',
3404 delim
=', ') or 'none'
3405 write_debug('Optional libraries: %s' % lib_str
)
3408 for handler
in self
._opener
.handlers
:
3409 if hasattr(handler
, 'proxies'):
3410 proxy_map
.update(handler
.proxies
)
3411 write_debug(f
'Proxy map: {proxy_map}')
3414 if False and self
.params
.get('call_home'):
3415 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3416 write_debug('Public IP address: %s' % ipaddr
)
3417 latest_version
= self
.urlopen(
3418 'https://yt-dl.org/latest/version').read().decode('utf-8')
3419 if version_tuple(latest_version
) > version_tuple(__version__
):
3420 self
.report_warning(
3421 'You are using an outdated version (newest version: %s)! '
3422 'See https://yt-dl.org/update if you need help updating.' %
3425 def _setup_opener(self
):
3426 timeout_val
= self
.params
.get('socket_timeout')
3427 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3429 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3430 opts_cookiefile
= self
.params
.get('cookiefile')
3431 opts_proxy
= self
.params
.get('proxy')
3433 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3435 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3436 if opts_proxy
is not None:
3437 if opts_proxy
== '':
3440 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3442 proxies
= compat_urllib_request
.getproxies()
3443 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3444 if 'http' in proxies
and 'https' not in proxies
:
3445 proxies
['https'] = proxies
['http']
3446 proxy_handler
= PerRequestProxyHandler(proxies
)
3448 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3449 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3450 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3451 redirect_handler
= YoutubeDLRedirectHandler()
3452 data_handler
= compat_urllib_request_DataHandler()
3454 # When passing our own FileHandler instance, build_opener won't add the
3455 # default FileHandler and allows us to disable the file protocol, which
3456 # can be used for malicious purposes (see
3457 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3458 file_handler
= compat_urllib_request
.FileHandler()
3460 def file_open(*args
, **kwargs
):
3461 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3462 file_handler
.file_open
= file_open
3464 opener
= compat_urllib_request
.build_opener(
3465 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3467 # Delete the default user-agent header, which would otherwise apply in
3468 # cases where our custom HTTP handler doesn't come into play
3469 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3470 opener
.addheaders
= []
3471 self
._opener
= opener
3473 def encode(self
, s
):
3474 if isinstance(s
, bytes):
3475 return s
# Already encoded
3478 return s
.encode(self
.get_encoding())
3479 except UnicodeEncodeError as err
:
3480 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3483 def get_encoding(self
):
3484 encoding
= self
.params
.get('encoding')
3485 if encoding
is None:
3486 encoding
= preferredencoding()
3489 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3490 ''' Write infojson and returns True = written, False = skip, None = error '''
3491 if overwrite
is None:
3492 overwrite
= self
.params
.get('overwrites', True)
3493 if not self
.params
.get('writeinfojson'):
3496 self
.write_debug(f
'Skipping writing {label} infojson')
3498 elif not self
._ensure
_dir
_exists
(infofn
):
3500 elif not overwrite
and os
.path
.exists(infofn
):
3501 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3503 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3505 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3506 except (OSError, IOError):
3507 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3511 def _write_description(self
, label
, ie_result
, descfn
):
3512 ''' Write description and returns True = written, False = skip, None = error '''
3513 if not self
.params
.get('writedescription'):
3516 self
.write_debug(f
'Skipping writing {label} description')
3518 elif not self
._ensure
_dir
_exists
(descfn
):
3520 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3521 self
.to_screen(f
'[info] {label.title()} description is already present')
3522 elif ie_result
.get('description') is None:
3523 self
.report_warning(f
'There\'s no {label} description to write')
3527 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3528 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3529 descfile
.write(ie_result
['description'])
3530 except (OSError, IOError):
3531 self
.report_error(f
'Cannot write {label} description file {descfn}')
3535 def _write_subtitles(self
, info_dict
, filename
):
3536 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3538 subtitles
= info_dict
.get('requested_subtitles')
3539 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3540 # subtitles download errors are already managed as troubles in relevant IE
3541 # that way it will silently go on when used with unsupporting IE
3544 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3545 if not sub_filename_base
:
3546 self
.to_screen('[info] Skipping writing video subtitles')
3548 for sub_lang
, sub_info
in subtitles
.items():
3549 sub_format
= sub_info
['ext']
3550 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3551 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3552 if not self
.params
.get('overwrites', True) and os
.path
.exists(sub_filename
):
3553 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3554 sub_info
['filepath'] = sub_filename
3555 ret
.append((sub_filename
, sub_filename_final
))
3558 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3559 if sub_info
.get('data') is not None:
3561 # Use newline='' to prevent conversion of newline characters
3562 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3563 with io
.open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3564 subfile
.write(sub_info
['data'])
3565 sub_info
['filepath'] = sub_filename
3566 ret
.append((sub_filename
, sub_filename_final
))
3568 except (OSError, IOError):
3569 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3573 sub_copy
= sub_info
.copy()
3574 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3575 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3576 sub_info
['filepath'] = sub_filename
3577 ret
.append((sub_filename
, sub_filename_final
))
3578 except (ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3579 self
.report_warning(f
'Unable to download video subtitles for {sub_lang!r}: {err}')
3583 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3584 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3585 write_all
= self
.params
.get('write_all_thumbnails', False)
3586 thumbnails
, ret
= [], []
3587 if write_all
or self
.params
.get('writethumbnail', False):
3588 thumbnails
= info_dict
.get('thumbnails') or []
3589 multiple
= write_all
and len(thumbnails
) > 1
3591 if thumb_filename_base
is None:
3592 thumb_filename_base
= filename
3593 if thumbnails
and not thumb_filename_base
:
3594 self
.write_debug(f
'Skipping writing {label} thumbnail')
3597 for t
in thumbnails
[::-1]:
3598 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3599 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3600 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3601 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3603 if not self
.params
.get('overwrites', True) and os
.path
.exists(thumb_filename
):
3604 ret
.append((thumb_filename
, thumb_filename_final
))
3605 t
['filepath'] = thumb_filename
3606 self
.to_screen('[info] %s is already present' % (
3607 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3609 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3611 uf
= self
.urlopen(t
['url'])
3612 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3613 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3614 shutil
.copyfileobj(uf
, thumbf
)
3615 ret
.append((thumb_filename
, thumb_filename_final
))
3616 t
['filepath'] = thumb_filename
3617 except network_exceptions
as err
:
3618 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3619 if ret
and not write_all
: