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))
850 SUPPRESS
= 'light black'
852 def __format_text(self
, out
, text
, f
, fallback
=None, *, test_encoding
=False):
853 assert out
in ('screen', 'err')
856 handle
= self
._screen
_file
if out
== 'screen' else self
._err
_file
857 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', 'ascii')
858 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
859 if fallback
is not None and text
!= original_text
:
861 if isinstance(f
, self
.Styles
):
863 return format_text(text
, f
) if self
._allow
_colors
[out
] else text
if fallback
is None else fallback
865 def _format_screen(self
, *args
, **kwargs
):
866 return self
.__format
_text
('screen', *args
, **kwargs
)
868 def _format_err(self
, *args
, **kwargs
):
869 return self
.__format
_text
('err', *args
, **kwargs
)
871 def report_warning(self
, message
, only_once
=False):
873 Print the message to stderr, it will be prefixed with 'WARNING:'
874 If stderr is a tty file the 'WARNING:' will be colored
876 if self
.params
.get('logger') is not None:
877 self
.params
['logger'].warning(message
)
879 if self
.params
.get('no_warnings'):
881 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
883 def report_error(self
, message
, tb
=None):
885 Do the same as trouble, but prefixes the message with 'ERROR:', colored
886 in red if stderr is a tty file.
888 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', tb
)
890 def write_debug(self
, message
, only_once
=False):
891 '''Log debug message or Print message to stderr'''
892 if not self
.params
.get('verbose', False):
894 message
= '[debug] %s' % message
895 if self
.params
.get('logger'):
896 self
.params
['logger'].debug(message
)
898 self
.to_stderr(message
, only_once
)
900 def report_file_already_downloaded(self
, file_name
):
901 """Report file has already been fully downloaded."""
903 self
.to_screen('[download] %s has already been downloaded' % file_name
)
904 except UnicodeEncodeError:
905 self
.to_screen('[download] The file has already been downloaded')
907 def report_file_delete(self
, file_name
):
908 """Report that existing file will be deleted."""
910 self
.to_screen('Deleting existing file %s' % file_name
)
911 except UnicodeEncodeError:
912 self
.to_screen('Deleting existing file')
914 def raise_no_formats(self
, info
, forced
=False):
915 has_drm
= info
.get('__has_drm')
916 msg
= 'This video is DRM protected' if has_drm
else 'No video formats found!'
917 expected
= self
.params
.get('ignore_no_formats_error')
918 if forced
or not expected
:
919 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
920 expected
=has_drm
or expected
)
922 self
.report_warning(msg
)
924 def parse_outtmpl(self
):
925 outtmpl_dict
= self
.params
.get('outtmpl', {})
926 if not isinstance(outtmpl_dict
, dict):
927 outtmpl_dict
= {'default': outtmpl_dict}
928 # Remove spaces in the default template
929 if self
.params
.get('restrictfilenames'):
930 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
932 sanitize
= lambda x
: x
933 outtmpl_dict
.update({
934 k
: sanitize(v
) for k
, v
in DEFAULT_OUTTMPL
.items()
935 if outtmpl_dict
.get(k
) is None})
936 for key
, val
in outtmpl_dict
.items():
937 if isinstance(val
, bytes):
939 'Parameter outtmpl is bytes, but should be a unicode string. '
940 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
943 def get_output_path(self
, dir_type
='', filename
=None):
944 paths
= self
.params
.get('paths', {})
945 assert isinstance(paths
, dict)
947 expand_path(paths
.get('home', '').strip()),
948 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
951 # Temporary fix for #4787
952 # 'Treat' all problem characters by passing filename through preferredencoding
953 # to workaround encoding issues with subprocess on python2 @ Windows
954 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
955 path
= encodeFilename(path
, True).decode(preferredencoding())
956 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
959 def _outtmpl_expandpath(outtmpl
):
960 # expand_path translates '%%' into '%' and '$$' into '$'
961 # correspondingly that is not what we want since we need to keep
962 # '%%' intact for template dict substitution step. Working around
963 # with boundary-alike separator hack.
964 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
965 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
967 # outtmpl should be expand_path'ed before template dict substitution
968 # because meta fields may contain env variables we don't want to
969 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
970 # title "Hello $PATH", we don't want `$PATH` to be expanded.
971 return expand_path(outtmpl
).replace(sep
, '')
974 def escape_outtmpl(outtmpl
):
975 ''' Escape any remaining strings like %s, %abc% etc. '''
977 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
978 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
982 def validate_outtmpl(cls
, outtmpl
):
983 ''' @return None or Exception object '''
985 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljqBU]'),
986 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
987 cls
._outtmpl
_expandpath
(outtmpl
))
989 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
991 except ValueError as err
:
995 def _copy_infodict(info_dict
):
996 info_dict
= dict(info_dict
)
997 for key
in ('__original_infodict', '__postprocessors'):
998 info_dict
.pop(key
, None)
1001 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=None):
1002 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict """
1003 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1005 info_dict
= self
._copy
_infodict
(info_dict
)
1006 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1007 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1008 if info_dict
.get('duration', None) is not None
1010 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
1011 if info_dict
.get('resolution') is None:
1012 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1014 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1015 # of %(field)s to %(field)0Nd for backward compatibility
1016 field_size_compat_map
= {
1017 'playlist_index': number_of_digits(info_dict
.get('_last_playlist_index') or 0),
1018 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1019 'autonumber': self
.params
.get('autonumber_size') or 5,
1023 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljqBU]'))
1028 # Field is of the form key1.key2...
1029 # where keys (except first) can be string, int or slice
1030 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1031 MATH_FIELD_RE
= r
'''{field}|{num}'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
1032 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1033 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
1036 (?P<maths>(?:{math_op}{math_field})*)
1037 (?:>(?P<strf_format>.+?))?
1038 (?P<alternate>(?<!\\),[^|)]+)?
1039 (?:\|(?P<default>.*?))?
1040 $'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
1042 def _traverse_infodict(k
):
1046 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1048 def get_value(mdict
):
1050 value
= _traverse_infodict(mdict
['fields'])
1053 value
= float_or_none(value
)
1054 if value
is not None:
1057 offset_key
= mdict
['maths']
1059 value
= float_or_none(value
)
1063 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1064 offset_key
).group(0)
1065 offset_key
= offset_key
[len(item
):]
1066 if operator
is None:
1067 operator
= MATH_FUNCTIONS
[item
]
1069 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1070 offset
= float_or_none(item
)
1072 offset
= float_or_none(_traverse_infodict(item
))
1074 value
= operator(value
, multiplier
* offset
)
1075 except (TypeError, ZeroDivisionError):
1078 # Datetime formatting
1079 if mdict
['strf_format']:
1080 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1084 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1086 def _dumpjson_default(obj
):
1087 if isinstance(obj
, (set, LazyList
)):
1089 raise TypeError(f
'Object of type {type(obj).__name__} is not JSON serializable')
1091 def create_key(outer_mobj
):
1092 if not outer_mobj
.group('has_key'):
1093 return outer_mobj
.group(0)
1094 key
= outer_mobj
.group('key')
1095 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1096 initial_field
= mobj
.group('fields').split('.')[-1] if mobj
else ''
1097 value
, default
= None, na
1099 mobj
= mobj
.groupdict()
1100 default
= mobj
['default'] if mobj
['default'] is not None else default
1101 value
= get_value(mobj
)
1102 if value
is None and mobj
['alternate']:
1103 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['alternate'][1:])
1107 fmt
= outer_mobj
.group('format')
1108 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1109 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
1111 value
= default
if value
is None else value
1113 flags
= outer_mobj
.group('conversion') or ''
1114 str_fmt
= f
'{fmt[:-1]}s'
1115 if fmt
[-1] == 'l': # list
1116 delim
= '\n' if '#' in flags
else ', '
1117 value
, fmt
= delim
.join(variadic(value
)), str_fmt
1118 elif fmt
[-1] == 'j': # json
1119 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1120 elif fmt
[-1] == 'q': # quoted
1121 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1122 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1123 elif fmt
[-1] == 'B': # bytes
1124 value
= f
'%{str_fmt}'.encode('utf-8') % str(value
).encode('utf-8')
1125 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1126 elif fmt
[-1] == 'U': # unicode normalized
1127 value
, fmt
= unicodedata
.normalize(
1128 # "+" = compatibility equivalence, "#" = NFD
1129 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1131 elif fmt
[-1] == 'c':
1133 value
= str(value
)[0]
1136 elif fmt
[-1] not in 'rs': # numeric
1137 value
= float_or_none(value
)
1139 value
, fmt
= default
, 's'
1143 # If value is an object, sanitize might convert it to a string
1144 # So we convert it to repr first
1145 value
, fmt
= repr(value
), str_fmt
1146 if fmt
[-1] in 'csr':
1147 value
= sanitize(initial_field
, value
)
1149 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1150 TMPL_DICT
[key
] = value
1151 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1153 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1155 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1156 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1157 return self
.escape_outtmpl(outtmpl
) % info_dict
1159 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
1161 sanitize
= lambda k
, v
: sanitize_filename(
1163 restricted
=self
.params
.get('restrictfilenames'),
1164 is_id
=(k
== 'id' or k
.endswith('_id')))
1165 outtmpl
= self
._outtmpl
_expandpath
(self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default']))
1166 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, sanitize
)
1168 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
1169 if filename
and force_ext
is not None:
1170 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1172 # https://github.com/blackjack4494/youtube-dlc/issues/85
1173 trim_file_name
= self
.params
.get('trim_file_name', False)
1175 fn_groups
= filename
.rsplit('.')
1178 if len(fn_groups
) > 2:
1179 sub_ext
= fn_groups
[-2]
1180 filename
= join_nonempty(fn_groups
[0][:trim_file_name
], sub_ext
, ext
, delim
='.')
1183 except ValueError as err
:
1184 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1187 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
1188 """Generate the output filename."""
1190 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
1191 if not filename
and dir_type
not in ('', 'temp'):
1195 if not self
.params
.get('paths'):
1197 elif filename
== '-':
1198 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1199 elif os
.path
.isabs(filename
):
1200 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1201 if filename
== '-' or not filename
:
1204 return self
.get_output_path(dir_type
, filename
)
1206 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1207 """ Returns None if the file should be downloaded """
1209 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1212 if 'title' in info_dict
:
1213 # This can happen when we're just evaluating the playlist
1214 title
= info_dict
['title']
1215 matchtitle
= self
.params
.get('matchtitle', False)
1217 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1218 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1219 rejecttitle
= self
.params
.get('rejecttitle', False)
1221 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1222 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1223 date
= info_dict
.get('upload_date')
1224 if date
is not None:
1225 dateRange
= self
.params
.get('daterange', DateRange())
1226 if date
not in dateRange
:
1227 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1228 view_count
= info_dict
.get('view_count')
1229 if view_count
is not None:
1230 min_views
= self
.params
.get('min_views')
1231 if min_views
is not None and view_count
< min_views
:
1232 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1233 max_views
= self
.params
.get('max_views')
1234 if max_views
is not None and view_count
> max_views
:
1235 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1236 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1237 return 'Skipping "%s" because it is age restricted' % video_title
1239 match_filter
= self
.params
.get('match_filter')
1240 if match_filter
is not None:
1242 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1244 # For backward compatibility
1245 ret
= None if incomplete
else match_filter(info_dict
)
1250 if self
.in_download_archive(info_dict
):
1251 reason
= '%s has already been recorded in the archive' % video_title
1252 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1254 reason
= check_filter()
1255 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1256 if reason
is not None:
1258 self
.to_screen('[download] ' + reason
)
1259 if self
.params
.get(break_opt
, False):
1264 def add_extra_info(info_dict
, extra_info
):
1265 '''Set the keys from extra_info in info dict if they are missing'''
1266 for key
, value
in extra_info
.items():
1267 info_dict
.setdefault(key
, value
)
1269 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1270 process
=True, force_generic_extractor
=False):
1272 Return a list with a dictionary for each video extracted.
1275 url -- URL to extract
1278 download -- whether to download videos during extraction
1279 ie_key -- extractor key hint
1280 extra_info -- dictionary containing the extra values to add to each result
1281 process -- whether to resolve all unresolved references (URLs, playlist items),
1282 must be True for download to work.
1283 force_generic_extractor -- force using the generic extractor
1286 if extra_info
is None:
1289 if not ie_key
and force_generic_extractor
:
1293 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1297 for ie_key
, ie
in ies
.items():
1298 if not ie
.suitable(url
):
1301 if not ie
.working():
1302 self
.report_warning('The program functionality for this site has been marked as broken, '
1303 'and will probably not work.')
1305 temp_id
= ie
.get_temp_id(url
)
1306 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1307 self
.to_screen("[%s] %s: has already been recorded in archive" % (
1310 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1312 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1314 def __handle_extraction_exceptions(func
):
1315 @functools.wraps(func
)
1316 def wrapper(self
, *args
, **kwargs
):
1318 return func(self
, *args
, **kwargs
)
1319 except GeoRestrictedError
as e
:
1322 msg
+= '\nThis video is available in %s.' % ', '.join(
1323 map(ISO3166Utils
.short2full
, e
.countries
))
1324 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1325 self
.report_error(msg
)
1326 except ExtractorError
as e
: # An error we somewhat expected
1327 self
.report_error(compat_str(e
), e
.format_traceback())
1328 except ThrottledDownload
as e
:
1329 self
.to_stderr('\r')
1330 self
.report_warning(f
'{e}; Re-extracting data')
1331 return wrapper(self
, *args
, **kwargs
)
1332 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1334 except Exception as e
:
1335 if self
.params
.get('ignoreerrors'):
1336 self
.report_error(error_to_compat_str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1341 @__handle_extraction_exceptions
1342 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1343 ie_result
= ie
.extract(url
)
1344 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1346 if isinstance(ie_result
, list):
1347 # Backwards compatibility: old IE result format
1349 '_type': 'compat_list',
1350 'entries': ie_result
,
1352 if extra_info
.get('original_url'):
1353 ie_result
.setdefault('original_url', extra_info
['original_url'])
1354 self
.add_default_extra_info(ie_result
, ie
, url
)
1356 return self
.process_ie_result(ie_result
, download
, extra_info
)
1360 def add_default_extra_info(self
, ie_result
, ie
, url
):
1362 self
.add_extra_info(ie_result
, {
1364 'original_url': url
,
1365 'webpage_url_basename': url_basename(url
),
1368 self
.add_extra_info(ie_result
, {
1369 'extractor': ie
.IE_NAME
,
1370 'extractor_key': ie
.ie_key(),
1373 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1375 Take the result of the ie(may be modified) and resolve all unresolved
1376 references (URLs, playlist items).
1378 It will also download the videos if 'download'.
1379 Returns the resolved ie_result.
1381 if extra_info
is None:
1383 result_type
= ie_result
.get('_type', 'video')
1385 if result_type
in ('url', 'url_transparent'):
1386 ie_result
['url'] = sanitize_url(ie_result
['url'])
1387 if ie_result
.get('original_url'):
1388 extra_info
.setdefault('original_url', ie_result
['original_url'])
1390 extract_flat
= self
.params
.get('extract_flat', False)
1391 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1392 or extract_flat
is True):
1393 info_copy
= ie_result
.copy()
1394 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1395 if ie
and not ie_result
.get('id'):
1396 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1397 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1398 self
.add_extra_info(info_copy
, extra_info
)
1399 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1400 if self
.params
.get('force_write_download_archive', False):
1401 self
.record_download_archive(info_copy
)
1404 if result_type
== 'video':
1405 self
.add_extra_info(ie_result
, extra_info
)
1406 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1407 additional_urls
= (ie_result
or {}).get('additional_urls')
1409 # TODO: Improve MetadataParserPP to allow setting a list
1410 if isinstance(additional_urls
, compat_str
):
1411 additional_urls
= [additional_urls
]
1413 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1414 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1415 ie_result
['additional_entries'] = [
1417 url
, download
, extra_info
,
1418 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1419 for url
in additional_urls
1422 elif result_type
== 'url':
1423 # We have to add extra_info to the results because it may be
1424 # contained in a playlist
1425 return self
.extract_info(
1426 ie_result
['url'], download
,
1427 ie_key
=ie_result
.get('ie_key'),
1428 extra_info
=extra_info
)
1429 elif result_type
== 'url_transparent':
1430 # Use the information from the embedding page
1431 info
= self
.extract_info(
1432 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1433 extra_info
=extra_info
, download
=False, process
=False)
1435 # extract_info may return None when ignoreerrors is enabled and
1436 # extraction failed with an error, don't crash and return early
1441 force_properties
= dict(
1442 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1443 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1444 if f
in force_properties
:
1445 del force_properties
[f
]
1446 new_result
= info
.copy()
1447 new_result
.update(force_properties
)
1449 # Extracted info may not be a video result (i.e.
1450 # info.get('_type', 'video') != video) but rather an url or
1451 # url_transparent. In such cases outer metadata (from ie_result)
1452 # should be propagated to inner one (info). For this to happen
1453 # _type of info should be overridden with url_transparent. This
1454 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1455 if new_result
.get('_type') == 'url':
1456 new_result
['_type'] = 'url_transparent'
1458 return self
.process_ie_result(
1459 new_result
, download
=download
, extra_info
=extra_info
)
1460 elif result_type
in ('playlist', 'multi_video'):
1461 # Protect from infinite recursion due to recursively nested playlists
1462 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1463 webpage_url
= ie_result
['webpage_url']
1464 if webpage_url
in self
._playlist
_urls
:
1466 '[download] Skipping already downloaded playlist: %s'
1467 % ie_result
.get('title') or ie_result
.get('id'))
1470 self
._playlist
_level
+= 1
1471 self
._playlist
_urls
.add(webpage_url
)
1472 self
._sanitize
_thumbnails
(ie_result
)
1474 return self
.__process
_playlist
(ie_result
, download
)
1476 self
._playlist
_level
-= 1
1477 if not self
._playlist
_level
:
1478 self
._playlist
_urls
.clear()
1479 elif result_type
== 'compat_list':
1480 self
.report_warning(
1481 'Extractor %s returned a compat_list result. '
1482 'It needs to be updated.' % ie_result
.get('extractor'))
1485 self
.add_extra_info(r
, {
1486 'extractor': ie_result
['extractor'],
1487 'webpage_url': ie_result
['webpage_url'],
1488 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1489 'extractor_key': ie_result
['extractor_key'],
1492 ie_result
['entries'] = [
1493 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1494 for r
in ie_result
['entries']
1498 raise Exception('Invalid result type: %s' % result_type
)
1500 def _ensure_dir_exists(self
, path
):
1501 return make_dir(path
, self
.report_error
)
1503 def __process_playlist(self
, ie_result
, download
):
1504 # We process each entry in the playlist
1505 playlist
= ie_result
.get('title') or ie_result
.get('id')
1506 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1508 if 'entries' not in ie_result
:
1509 raise EntryNotInPlaylist('There are no entries')
1511 MissingEntry
= object()
1512 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1513 if incomplete_entries
:
1514 def fill_missing_entries(entries
, indices
):
1515 ret
= [MissingEntry
] * max(indices
)
1516 for i
, entry
in zip(indices
, entries
):
1519 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1521 playlist_results
= []
1523 playliststart
= self
.params
.get('playliststart', 1)
1524 playlistend
= self
.params
.get('playlistend')
1525 # For backwards compatibility, interpret -1 as whole list
1526 if playlistend
== -1:
1529 playlistitems_str
= self
.params
.get('playlist_items')
1530 playlistitems
= None
1531 if playlistitems_str
is not None:
1532 def iter_playlistitems(format
):
1533 for string_segment
in format
.split(','):
1534 if '-' in string_segment
:
1535 start
, end
= string_segment
.split('-')
1536 for item
in range(int(start
), int(end
) + 1):
1539 yield int(string_segment
)
1540 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1542 ie_entries
= ie_result
['entries']
1544 'Downloading %d videos' if not isinstance(ie_entries
, list)
1545 else 'Collected %d videos; downloading %%d of them' % len(ie_entries
))
1547 if isinstance(ie_entries
, list):
1549 return ie_entries
[i
- 1]
1551 if not isinstance(ie_entries
, (PagedList
, LazyList
)):
1552 ie_entries
= LazyList(ie_entries
)
1555 return YoutubeDL
.__handle
_extraction
_exceptions
(
1556 lambda self
, i
: ie_entries
[i
- 1]
1560 items
= playlistitems
if playlistitems
is not None else itertools
.count(playliststart
)
1564 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1568 entry
= get_entry(i
)
1569 if entry
is MissingEntry
:
1570 raise EntryNotInPlaylist()
1571 except (IndexError, EntryNotInPlaylist
):
1572 if incomplete_entries
:
1573 raise EntryNotInPlaylist(f
'Entry {i} cannot be found')
1574 elif not playlistitems
:
1576 entries
.append(entry
)
1578 if entry
is not None:
1579 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1580 except (ExistingVideoReached
, RejectedVideoReached
):
1582 ie_result
['entries'] = entries
1584 # Save playlist_index before re-ordering
1586 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1587 for i
, entry
in enumerate(entries
, 1)
1588 if entry
is not None]
1589 n_entries
= len(entries
)
1591 if not playlistitems
and (playliststart
!= 1 or playlistend
):
1592 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1593 ie_result
['requested_entries'] = playlistitems
1595 _infojson_written
= False
1596 if not self
.params
.get('simulate') and self
.params
.get('allow_playlist_files', True):
1598 'playlist': playlist
,
1599 'playlist_id': ie_result
.get('id'),
1600 'playlist_title': ie_result
.get('title'),
1601 'playlist_uploader': ie_result
.get('uploader'),
1602 'playlist_uploader_id': ie_result
.get('uploader_id'),
1603 'playlist_index': 0,
1604 'n_entries': n_entries
,
1606 ie_copy
.update(dict(ie_result
))
1608 _infojson_written
= self
._write
_info
_json
(
1609 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1610 if _infojson_written
is None:
1612 if self
._write
_description
('playlist', ie_result
,
1613 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1615 # TODO: This should be passed to ThumbnailsConvertor if necessary
1616 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1618 if self
.params
.get('playlistreverse', False):
1619 entries
= entries
[::-1]
1620 if self
.params
.get('playlistrandom', False):
1621 random
.shuffle(entries
)
1623 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1625 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1627 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1628 for i
, entry_tuple
in enumerate(entries
, 1):
1629 playlist_index
, entry
= entry_tuple
1630 if 'playlist-index' in self
.params
.get('compat_opts', []):
1631 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1
1632 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1633 # This __x_forwarded_for_ip thing is a bit ugly but requires
1636 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1638 'n_entries': n_entries
,
1639 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1640 'playlist_index': playlist_index
,
1641 'playlist_autonumber': i
,
1642 'playlist': playlist
,
1643 'playlist_id': ie_result
.get('id'),
1644 'playlist_title': ie_result
.get('title'),
1645 'playlist_uploader': ie_result
.get('uploader'),
1646 'playlist_uploader_id': ie_result
.get('uploader_id'),
1647 'extractor': ie_result
['extractor'],
1648 'webpage_url': ie_result
['webpage_url'],
1649 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1650 'extractor_key': ie_result
['extractor_key'],
1653 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1656 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1657 if not entry_result
:
1659 if failures
>= max_failures
:
1661 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1663 playlist_results
.append(entry_result
)
1664 ie_result
['entries'] = playlist_results
1666 # Write the updated info to json
1667 if _infojson_written
and self
._write
_info
_json
(
1668 'updated playlist', ie_result
,
1669 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1671 self
.to_screen('[download] Finished downloading playlist: %s' % playlist
)
1674 @__handle_extraction_exceptions
1675 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1676 return self
.process_ie_result(
1677 entry
, download
=download
, extra_info
=extra_info
)
1679 def _build_format_filter(self
, filter_spec
):
1680 " Returns a function to filter the formats according to the filter_spec "
1690 operator_rex
= re
.compile(r
'''(?x)\s*
1691 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1692 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1693 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1694 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1695 m
= operator_rex
.fullmatch(filter_spec
)
1698 comparison_value
= int(m
.group('value'))
1700 comparison_value
= parse_filesize(m
.group('value'))
1701 if comparison_value
is None:
1702 comparison_value
= parse_filesize(m
.group('value') + 'B')
1703 if comparison_value
is None:
1705 'Invalid value %r in format specification %r' % (
1706 m
.group('value'), filter_spec
))
1707 op
= OPERATORS
[m
.group('op')]
1712 '^=': lambda attr
, value
: attr
.startswith(value
),
1713 '$=': lambda attr
, value
: attr
.endswith(value
),
1714 '*=': lambda attr
, value
: value
in attr
,
1716 str_operator_rex
= re
.compile(r
'''(?x)\s*
1717 (?P<key>[a-zA-Z0-9._-]+)\s*
1718 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1719 (?P<value>[a-zA-Z0-9._-]+)\s*
1720 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1721 m
= str_operator_rex
.fullmatch(filter_spec
)
1723 comparison_value
= m
.group('value')
1724 str_op
= STR_OPERATORS
[m
.group('op')]
1725 if m
.group('negation'):
1726 op
= lambda attr
, value
: not str_op(attr
, value
)
1731 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1734 actual_value
= f
.get(m
.group('key'))
1735 if actual_value
is None:
1736 return m
.group('none_inclusive')
1737 return op(actual_value
, comparison_value
)
1740 def _check_formats(self
, formats
):
1742 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1743 path
= self
.get_output_path('temp')
1744 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1746 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1749 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1750 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1753 if os
.path
.exists(temp_file
.name
):
1755 os
.remove(temp_file
.name
)
1757 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1761 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1763 def _default_format_spec(self
, info_dict
, download
=True):
1766 merger
= FFmpegMergerPP(self
)
1767 return merger
.available
and merger
.can_merge()
1770 not self
.params
.get('simulate')
1774 or info_dict
.get('is_live', False)
1775 or self
.outtmpl_dict
['default'] == '-'))
1778 or self
.params
.get('allow_multiple_audio_streams', False)
1779 or 'format-spec' in self
.params
.get('compat_opts', []))
1782 'best/bestvideo+bestaudio' if prefer_best
1783 else 'bestvideo*+bestaudio/best' if not compat
1784 else 'bestvideo+bestaudio/best')
1786 def build_format_selector(self
, format_spec
):
1787 def syntax_error(note
, start
):
1789 'Invalid format specification: '
1790 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1791 return SyntaxError(message
)
1793 PICKFIRST
= 'PICKFIRST'
1797 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1799 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1800 'video': self
.params
.get('allow_multiple_video_streams', False)}
1802 check_formats
= self
.params
.get('check_formats') == 'selected'
1804 def _parse_filter(tokens
):
1806 for type, string
, start
, _
, _
in tokens
:
1807 if type == tokenize
.OP
and string
== ']':
1808 return ''.join(filter_parts
)
1810 filter_parts
.append(string
)
1812 def _remove_unused_ops(tokens
):
1813 # Remove operators that we don't use and join them with the surrounding strings
1814 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1815 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1816 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1817 for type, string
, start
, end
, line
in tokens
:
1818 if type == tokenize
.OP
and string
== '[':
1820 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1822 yield type, string
, start
, end
, line
1823 # everything inside brackets will be handled by _parse_filter
1824 for type, string
, start
, end
, line
in tokens
:
1825 yield type, string
, start
, end
, line
1826 if type == tokenize
.OP
and string
== ']':
1828 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1830 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1832 yield type, string
, start
, end
, line
1833 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1835 last_string
= string
1839 last_string
+= string
1841 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1843 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1845 current_selector
= None
1846 for type, string
, start
, _
, _
in tokens
:
1847 # ENCODING is only defined in python 3.x
1848 if type == getattr(tokenize
, 'ENCODING', None):
1850 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1851 current_selector
= FormatSelector(SINGLE
, string
, [])
1852 elif type == tokenize
.OP
:
1854 if not inside_group
:
1855 # ')' will be handled by the parentheses group
1856 tokens
.restore_last_token()
1858 elif inside_merge
and string
in ['/', ',']:
1859 tokens
.restore_last_token()
1861 elif inside_choice
and string
== ',':
1862 tokens
.restore_last_token()
1865 if not current_selector
:
1866 raise syntax_error('"," must follow a format selector', start
)
1867 selectors
.append(current_selector
)
1868 current_selector
= None
1870 if not current_selector
:
1871 raise syntax_error('"/" must follow a format selector', start
)
1872 first_choice
= current_selector
1873 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1874 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1876 if not current_selector
:
1877 current_selector
= FormatSelector(SINGLE
, 'best', [])
1878 format_filter
= _parse_filter(tokens
)
1879 current_selector
.filters
.append(format_filter
)
1881 if current_selector
:
1882 raise syntax_error('Unexpected "("', start
)
1883 group
= _parse_format_selection(tokens
, inside_group
=True)
1884 current_selector
= FormatSelector(GROUP
, group
, [])
1886 if not current_selector
:
1887 raise syntax_error('Unexpected "+"', start
)
1888 selector_1
= current_selector
1889 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1891 raise syntax_error('Expected a selector', start
)
1892 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
1894 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
1895 elif type == tokenize
.ENDMARKER
:
1897 if current_selector
:
1898 selectors
.append(current_selector
)
1901 def _merge(formats_pair
):
1902 format_1
, format_2
= formats_pair
1905 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
1906 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
1908 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
1909 get_no_more
= {'video': False, 'audio': False}
1910 for (i
, fmt_info
) in enumerate(formats_info
):
1911 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
1914 for aud_vid
in ['audio', 'video']:
1915 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
1916 if get_no_more
[aud_vid
]:
1919 get_no_more
[aud_vid
] = True
1921 if len(formats_info
) == 1:
1922 return formats_info
[0]
1924 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
1925 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
1927 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
1928 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
1930 output_ext
= self
.params
.get('merge_output_format')
1933 output_ext
= the_only_video
['ext']
1934 elif the_only_audio
and not video_fmts
:
1935 output_ext
= the_only_audio
['ext']
1939 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
1942 'requested_formats': formats_info
,
1943 'format': '+'.join(filtered('format')),
1944 'format_id': '+'.join(filtered('format_id')),
1946 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
1947 'language': '+'.join(orderedSet(filtered('language'))) or None,
1948 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
1949 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
1950 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
1955 'width': the_only_video
.get('width'),
1956 'height': the_only_video
.get('height'),
1957 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
1958 'fps': the_only_video
.get('fps'),
1959 'dynamic_range': the_only_video
.get('dynamic_range'),
1960 'vcodec': the_only_video
.get('vcodec'),
1961 'vbr': the_only_video
.get('vbr'),
1962 'stretched_ratio': the_only_video
.get('stretched_ratio'),
1967 'acodec': the_only_audio
.get('acodec'),
1968 'abr': the_only_audio
.get('abr'),
1969 'asr': the_only_audio
.get('asr'),
1974 def _check_formats(formats
):
1975 if not check_formats
:
1978 yield from self
._check
_formats
(formats
)
1980 def _build_selector_function(selector
):
1981 if isinstance(selector
, list): # ,
1982 fs
= [_build_selector_function(s
) for s
in selector
]
1984 def selector_function(ctx
):
1987 return selector_function
1989 elif selector
.type == GROUP
: # ()
1990 selector_function
= _build_selector_function(selector
.selector
)
1992 elif selector
.type == PICKFIRST
: # /
1993 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
1995 def selector_function(ctx
):
1997 picked_formats
= list(f(ctx
))
1999 return picked_formats
2002 elif selector
.type == MERGE
: # +
2003 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2005 def selector_function(ctx
):
2006 for pair
in itertools
.product(
2007 selector_1(copy
.deepcopy(ctx
)), selector_2(copy
.deepcopy(ctx
))):
2010 elif selector
.type == SINGLE
: # atom
2011 format_spec
= selector
.selector
or 'best'
2013 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2014 if format_spec
== 'all':
2015 def selector_function(ctx
):
2016 yield from _check_formats(ctx
['formats'][::-1])
2017 elif format_spec
== 'mergeall':
2018 def selector_function(ctx
):
2019 formats
= list(_check_formats(ctx
['formats']))
2022 merged_format
= formats
[-1]
2023 for f
in formats
[-2::-1]:
2024 merged_format
= _merge((merged_format
, f
))
2028 format_fallback
, format_reverse
, format_idx
= False, True, 1
2030 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2032 if mobj
is not None:
2033 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2034 format_reverse
= mobj
.group('bw')[0] == 'b'
2035 format_type
= (mobj
.group('type') or [None])[0]
2036 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2037 format_modified
= mobj
.group('mod') is not None
2039 format_fallback
= not format_type
and not format_modified
# for b, w
2041 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2042 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2043 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2044 if format_type
# bv, ba, wv, wa
2045 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2046 if not format_modified
# b, w
2047 else lambda f
: True) # b*, w*
2048 filter_f
= lambda f
: _filter_f(f
) and (
2049 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2051 if format_spec
in self
._format
_selection
_exts
['audio']:
2052 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2053 elif format_spec
in self
._format
_selection
_exts
['video']:
2054 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2055 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2056 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2058 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2060 def selector_function(ctx
):
2061 formats
= list(ctx
['formats'])
2062 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2063 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
2064 # for extractors with incomplete formats (audio only (soundcloud)
2065 # or video only (imgur)) best/worst will fallback to
2066 # best/worst {video,audio}-only format
2068 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2070 yield matches
[format_idx
- 1]
2074 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2076 def final_selector(ctx
):
2077 ctx_copy
= copy
.deepcopy(ctx
)
2078 for _filter
in filters
:
2079 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2080 return selector_function(ctx_copy
)
2081 return final_selector
2083 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
2085 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
2086 except tokenize
.TokenError
:
2087 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2089 class TokenIterator(object):
2090 def __init__(self
, tokens
):
2091 self
.tokens
= tokens
2098 if self
.counter
>= len(self
.tokens
):
2099 raise StopIteration()
2100 value
= self
.tokens
[self
.counter
]
2106 def restore_last_token(self
):
2109 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2110 return _build_selector_function(parsed_selector
)
2112 def _calc_headers(self
, info_dict
):
2113 res
= std_headers
.copy()
2115 add_headers
= info_dict
.get('http_headers')
2117 res
.update(add_headers
)
2119 cookies
= self
._calc
_cookies
(info_dict
)
2121 res
['Cookie'] = cookies
2123 if 'X-Forwarded-For' not in res
:
2124 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2125 if x_forwarded_for_ip
:
2126 res
['X-Forwarded-For'] = x_forwarded_for_ip
2130 def _calc_cookies(self
, info_dict
):
2131 pr
= sanitized_Request(info_dict
['url'])
2132 self
.cookiejar
.add_cookie_header(pr
)
2133 return pr
.get_header('Cookie')
2135 def _sort_thumbnails(self
, thumbnails
):
2136 thumbnails
.sort(key
=lambda t
: (
2137 t
.get('preference') if t
.get('preference') is not None else -1,
2138 t
.get('width') if t
.get('width') is not None else -1,
2139 t
.get('height') if t
.get('height') is not None else -1,
2140 t
.get('id') if t
.get('id') is not None else '',
2143 def _sanitize_thumbnails(self
, info_dict
):
2144 thumbnails
= info_dict
.get('thumbnails')
2145 if thumbnails
is None:
2146 thumbnail
= info_dict
.get('thumbnail')
2148 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2152 def check_thumbnails(thumbnails
):
2153 for t
in thumbnails
:
2154 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2156 self
.urlopen(HEADRequest(t
['url']))
2157 except network_exceptions
as err
:
2158 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2162 self
._sort
_thumbnails
(thumbnails
)
2163 for i
, t
in enumerate(thumbnails
):
2164 if t
.get('id') is None:
2166 if t
.get('width') and t
.get('height'):
2167 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2168 t
['url'] = sanitize_url(t
['url'])
2170 if self
.params
.get('check_formats') is True:
2171 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2173 info_dict
['thumbnails'] = thumbnails
2175 def process_video_result(self
, info_dict
, download
=True):
2176 assert info_dict
.get('_type', 'video') == 'video'
2178 if 'id' not in info_dict
:
2179 raise ExtractorError('Missing "id" field in extractor result')
2180 if 'title' not in info_dict
:
2181 raise ExtractorError('Missing "title" field in extractor result',
2182 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2184 def report_force_conversion(field
, field_not
, conversion
):
2185 self
.report_warning(
2186 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2187 % (field
, field_not
, conversion
))
2189 def sanitize_string_field(info
, string_field
):
2190 field
= info
.get(string_field
)
2191 if field
is None or isinstance(field
, compat_str
):
2193 report_force_conversion(string_field
, 'a string', 'string')
2194 info
[string_field
] = compat_str(field
)
2196 def sanitize_numeric_fields(info
):
2197 for numeric_field
in self
._NUMERIC
_FIELDS
:
2198 field
= info
.get(numeric_field
)
2199 if field
is None or isinstance(field
, compat_numeric_types
):
2201 report_force_conversion(numeric_field
, 'numeric', 'int')
2202 info
[numeric_field
] = int_or_none(field
)
2204 sanitize_string_field(info_dict
, 'id')
2205 sanitize_numeric_fields(info_dict
)
2207 if 'playlist' not in info_dict
:
2208 # It isn't part of a playlist
2209 info_dict
['playlist'] = None
2210 info_dict
['playlist_index'] = None
2212 self
._sanitize
_thumbnails
(info_dict
)
2214 thumbnail
= info_dict
.get('thumbnail')
2215 thumbnails
= info_dict
.get('thumbnails')
2217 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2219 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2221 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2222 info_dict
['display_id'] = info_dict
['id']
2224 if info_dict
.get('duration') is not None:
2225 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2227 for ts_key
, date_key
in (
2228 ('timestamp', 'upload_date'),
2229 ('release_timestamp', 'release_date'),
2231 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2232 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2233 # see http://bugs.python.org/issue1646728)
2235 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2236 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2237 except (ValueError, OverflowError, OSError):
2240 live_keys
= ('is_live', 'was_live')
2241 live_status
= info_dict
.get('live_status')
2242 if live_status
is None:
2243 for key
in live_keys
:
2244 if info_dict
.get(key
) is False:
2246 if info_dict
.get(key
):
2249 if all(info_dict
.get(key
) is False for key
in live_keys
):
2250 live_status
= 'not_live'
2252 info_dict
['live_status'] = live_status
2253 for key
in live_keys
:
2254 if info_dict
.get(key
) is None:
2255 info_dict
[key
] = (live_status
== key
)
2257 # Auto generate title fields corresponding to the *_number fields when missing
2258 # in order to always have clean titles. This is very common for TV series.
2259 for field
in ('chapter', 'season', 'episode'):
2260 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2261 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2263 for cc_kind
in ('subtitles', 'automatic_captions'):
2264 cc
= info_dict
.get(cc_kind
)
2266 for _
, subtitle
in cc
.items():
2267 for subtitle_format
in subtitle
:
2268 if subtitle_format
.get('url'):
2269 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2270 if subtitle_format
.get('ext') is None:
2271 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2273 automatic_captions
= info_dict
.get('automatic_captions')
2274 subtitles
= info_dict
.get('subtitles')
2276 info_dict
['requested_subtitles'] = self
.process_subtitles(
2277 info_dict
['id'], subtitles
, automatic_captions
)
2279 if info_dict
.get('formats') is None:
2280 # There's only one format available
2281 formats
= [info_dict
]
2283 formats
= info_dict
['formats']
2285 info_dict
['__has_drm'] = any(f
.get('has_drm') for f
in formats
)
2286 if not self
.params
.get('allow_unplayable_formats'):
2287 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2290 self
.raise_no_formats(info_dict
)
2292 def is_wellformed(f
):
2295 self
.report_warning(
2296 '"url" field is missing or empty - skipping format, '
2297 'there is an error in extractor')
2299 if isinstance(url
, bytes):
2300 sanitize_string_field(f
, 'url')
2303 # Filter out malformed formats for better extraction robustness
2304 formats
= list(filter(is_wellformed
, formats
))
2308 # We check that all the formats have the format and format_id fields
2309 for i
, format
in enumerate(formats
):
2310 sanitize_string_field(format
, 'format_id')
2311 sanitize_numeric_fields(format
)
2312 format
['url'] = sanitize_url(format
['url'])
2313 if not format
.get('format_id'):
2314 format
['format_id'] = compat_str(i
)
2316 # Sanitize format_id from characters used in format selector expression
2317 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2318 format_id
= format
['format_id']
2319 if format_id
not in formats_dict
:
2320 formats_dict
[format_id
] = []
2321 formats_dict
[format_id
].append(format
)
2323 # Make sure all formats have unique format_id
2324 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2325 for format_id
, ambiguous_formats
in formats_dict
.items():
2326 ambigious_id
= len(ambiguous_formats
) > 1
2327 for i
, format
in enumerate(ambiguous_formats
):
2329 format
['format_id'] = '%s-%d' % (format_id
, i
)
2330 if format
.get('ext') is None:
2331 format
['ext'] = determine_ext(format
['url']).lower()
2332 # Ensure there is no conflict between id and ext in format selection
2333 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2334 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2335 format
['format_id'] = 'f%s' % format
['format_id']
2337 for i
, format
in enumerate(formats
):
2338 if format
.get('format') is None:
2339 format
['format'] = '{id} - {res}{note}'.format(
2340 id=format
['format_id'],
2341 res
=self
.format_resolution(format
),
2342 note
=format_field(format
, 'format_note', ' (%s)'),
2344 if format
.get('protocol') is None:
2345 format
['protocol'] = determine_protocol(format
)
2346 if format
.get('resolution') is None:
2347 format
['resolution'] = self
.format_resolution(format
, default
=None)
2348 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2349 format
['dynamic_range'] = 'SDR'
2350 if (info_dict
.get('duration') and format
.get('tbr')
2351 and not format
.get('filesize') and not format
.get('filesize_approx')):
2352 format
['filesize_approx'] = info_dict
['duration'] * format
['tbr'] * (1024 / 8)
2354 # Add HTTP headers, so that external programs can use them from the
2356 full_format_info
= info_dict
.copy()
2357 full_format_info
.update(format
)
2358 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2359 # Remove private housekeeping stuff
2360 if '__x_forwarded_for_ip' in info_dict
:
2361 del info_dict
['__x_forwarded_for_ip']
2363 # TODO Central sorting goes here
2365 if self
.params
.get('check_formats') is True:
2366 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2368 if not formats
or formats
[0] is not info_dict
:
2369 # only set the 'formats' fields if the original info_dict list them
2370 # otherwise we end up with a circular reference, the first (and unique)
2371 # element in the 'formats' field in info_dict is info_dict itself,
2372 # which can't be exported to json
2373 info_dict
['formats'] = formats
2375 info_dict
, _
= self
.pre_process(info_dict
)
2377 # The pre-processors may have modified the formats
2378 formats
= info_dict
.get('formats', [info_dict
])
2380 if self
.params
.get('list_thumbnails'):
2381 self
.list_thumbnails(info_dict
)
2382 if self
.params
.get('listformats'):
2383 if not info_dict
.get('formats') and not info_dict
.get('url'):
2384 self
.to_screen('%s has no formats' % info_dict
['id'])
2386 self
.list_formats(info_dict
)
2387 if self
.params
.get('listsubtitles'):
2388 if 'automatic_captions' in info_dict
:
2389 self
.list_subtitles(
2390 info_dict
['id'], automatic_captions
, 'automatic captions')
2391 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2392 list_only
= self
.params
.get('simulate') is None and (
2393 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2395 # Without this printing, -F --print-json will not work
2396 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2399 format_selector
= self
.format_selector
2400 if format_selector
is None:
2401 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2402 self
.write_debug('Default format spec: %s' % req_format
)
2403 format_selector
= self
.build_format_selector(req_format
)
2405 # While in format selection we may need to have an access to the original
2406 # format set in order to calculate some metrics or do some processing.
2407 # For now we need to be able to guess whether original formats provided
2408 # by extractor are incomplete or not (i.e. whether extractor provides only
2409 # video-only or audio-only formats) for proper formats selection for
2410 # extractors with such incomplete formats (see
2411 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2412 # Since formats may be filtered during format selection and may not match
2413 # the original formats the results may be incorrect. Thus original formats
2414 # or pre-calculated metrics should be passed to format selection routines
2416 # We will pass a context object containing all necessary additional data
2417 # instead of just formats.
2418 # This fixes incorrect format selection issue (see
2419 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2420 incomplete_formats
= (
2421 # All formats are video-only or
2422 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2423 # all formats are audio-only
2424 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2428 'incomplete_formats': incomplete_formats
,
2431 formats_to_download
= list(format_selector(ctx
))
2432 if not formats_to_download
:
2433 if not self
.params
.get('ignore_no_formats_error'):
2434 raise ExtractorError('Requested format is not available', expected
=True,
2435 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2437 self
.report_warning('Requested format is not available')
2438 # Process what we can, even without any available formats.
2439 self
.process_info(dict(info_dict
))
2442 '[info] %s: Downloading %d format(s): %s' % (
2443 info_dict
['id'], len(formats_to_download
),
2444 ", ".join([f
['format_id'] for f
in formats_to_download
])))
2445 for fmt
in formats_to_download
:
2446 new_info
= dict(info_dict
)
2447 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2448 new_info
['__original_infodict'] = info_dict
2449 new_info
.update(fmt
)
2450 self
.process_info(new_info
)
2451 # We update the info dict with the selected best quality format (backwards compatibility)
2452 if formats_to_download
:
2453 info_dict
.update(formats_to_download
[-1])
2456 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2457 """Select the requested subtitles and their format"""
2459 if normal_subtitles
and self
.params
.get('writesubtitles'):
2460 available_subs
.update(normal_subtitles
)
2461 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2462 for lang
, cap_info
in automatic_captions
.items():
2463 if lang
not in available_subs
:
2464 available_subs
[lang
] = cap_info
2466 if (not self
.params
.get('writesubtitles') and not
2467 self
.params
.get('writeautomaticsub') or not
2471 all_sub_langs
= available_subs
.keys()
2472 if self
.params
.get('allsubtitles', False):
2473 requested_langs
= all_sub_langs
2474 elif self
.params
.get('subtitleslangs', False):
2475 # A list is used so that the order of languages will be the same as
2476 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2477 requested_langs
= []
2478 for lang_re
in self
.params
.get('subtitleslangs'):
2479 if lang_re
== 'all':
2480 requested_langs
.extend(all_sub_langs
)
2482 discard
= lang_re
[0] == '-'
2484 lang_re
= lang_re
[1:]
2485 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2487 for lang
in current_langs
:
2488 while lang
in requested_langs
:
2489 requested_langs
.remove(lang
)
2491 requested_langs
.extend(current_langs
)
2492 requested_langs
= orderedSet(requested_langs
)
2493 elif 'en' in available_subs
:
2494 requested_langs
= ['en']
2496 requested_langs
= [list(all_sub_langs
)[0]]
2498 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2500 formats_query
= self
.params
.get('subtitlesformat', 'best')
2501 formats_preference
= formats_query
.split('/') if formats_query
else []
2503 for lang
in requested_langs
:
2504 formats
= available_subs
.get(lang
)
2506 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2508 for ext
in formats_preference
:
2512 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2518 self
.report_warning(
2519 'No subtitle format found matching "%s" for language %s, '
2520 'using %s' % (formats_query
, lang
, f
['ext']))
2524 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2525 def print_mandatory(field
, actual_field
=None):
2526 if actual_field
is None:
2527 actual_field
= field
2528 if (self
.params
.get('force%s' % field
, False)
2529 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2530 self
.to_stdout(info_dict
[actual_field
])
2532 def print_optional(field
):
2533 if (self
.params
.get('force%s' % field
, False)
2534 and info_dict
.get(field
) is not None):
2535 self
.to_stdout(info_dict
[field
])
2537 info_dict
= info_dict
.copy()
2538 if filename
is not None:
2539 info_dict
['filename'] = filename
2540 if info_dict
.get('requested_formats') is not None:
2541 # For RTMP URLs, also include the playpath
2542 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2543 elif 'url' in info_dict
:
2544 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2546 if self
.params
.get('forceprint') or self
.params
.get('forcejson'):
2547 self
.post_extract(info_dict
)
2548 for tmpl
in self
.params
.get('forceprint', []):
2549 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2550 if mobj
and mobj
.group(1):
2551 tmpl
= f
'{tmpl[:-1]} = %({tmpl[:-1]})s'
2553 tmpl
= '%({})s'.format(tmpl
)
2554 self
.to_stdout(self
.evaluate_outtmpl(tmpl
, info_dict
))
2556 print_mandatory('title')
2557 print_mandatory('id')
2558 print_mandatory('url', 'urls')
2559 print_optional('thumbnail')
2560 print_optional('description')
2561 print_optional('filename')
2562 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2563 self
.to_stdout(formatSeconds(info_dict
['duration']))
2564 print_mandatory('format')
2566 if self
.params
.get('forcejson'):
2567 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2569 def dl(self
, name
, info
, subtitle
=False, test
=False):
2570 if not info
.get('url'):
2571 self
.raise_no_formats(info
, True)
2574 verbose
= self
.params
.get('verbose')
2577 'quiet': self
.params
.get('quiet') or not verbose
,
2579 'noprogress': not verbose
,
2581 'skip_unavailable_fragments': False,
2582 'keep_fragments': False,
2584 '_no_ytdl_file': True,
2587 params
= self
.params
2588 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2590 for ph
in self
._progress
_hooks
:
2591 fd
.add_progress_hook(ph
)
2592 urls
= '", "'.join([f
['url'] for f
in info
.get('requested_formats', [])] or [info
['url']])
2593 self
.write_debug('Invoking downloader on "%s"' % urls
)
2595 new_info
= copy
.deepcopy(self
._copy
_infodict
(info
))
2596 if new_info
.get('http_headers') is None:
2597 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2598 return fd
.download(name
, new_info
, subtitle
)
2600 def process_info(self
, info_dict
):
2601 """Process a single resolved IE result."""
2603 assert info_dict
.get('_type', 'video') == 'video'
2605 max_downloads
= self
.params
.get('max_downloads')
2606 if max_downloads
is not None:
2607 if self
._num
_downloads
>= int(max_downloads
):
2608 raise MaxDownloadsReached()
2610 # TODO: backward compatibility, to be removed
2611 info_dict
['fulltitle'] = info_dict
['title']
2613 if 'format' not in info_dict
and 'ext' in info_dict
:
2614 info_dict
['format'] = info_dict
['ext']
2616 if self
._match
_entry
(info_dict
) is not None:
2619 self
.post_extract(info_dict
)
2620 self
._num
_downloads
+= 1
2622 # info_dict['_filename'] needs to be set for backward compatibility
2623 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2624 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2628 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2630 if self
.params
.get('simulate'):
2631 if self
.params
.get('force_write_download_archive', False):
2632 self
.record_download_archive(info_dict
)
2633 # Do nothing else if in simulate mode
2636 if full_filename
is None:
2638 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2640 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2643 if self
._write
_description
('video', info_dict
,
2644 self
.prepare_filename(info_dict
, 'description')) is None:
2647 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2648 if sub_files
is None:
2650 files_to_move
.update(dict(sub_files
))
2652 thumb_files
= self
._write
_thumbnails
(
2653 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2654 if thumb_files
is None:
2656 files_to_move
.update(dict(thumb_files
))
2658 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2659 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2660 if _infojson_written
:
2661 info_dict
['infojson_filename'] = infofn
2662 # For backward compatability, even though it was a private field
2663 info_dict
['__infojson_filename'] = infofn
2664 elif _infojson_written
is None:
2667 # Note: Annotations are deprecated
2669 if self
.params
.get('writeannotations', False):
2670 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2672 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2674 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2675 self
.to_screen('[info] Video annotations are already present')
2676 elif not info_dict
.get('annotations'):
2677 self
.report_warning('There are no annotations to write.')
2680 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2681 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2682 annofile
.write(info_dict
['annotations'])
2683 except (KeyError, TypeError):
2684 self
.report_warning('There are no annotations to write.')
2685 except (OSError, IOError):
2686 self
.report_error('Cannot write annotations file: ' + annofn
)
2689 # Write internet shortcut files
2690 def _write_link_file(link_type
):
2691 if 'webpage_url' not in info_dict
:
2692 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2694 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2695 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
2697 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2698 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2701 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2702 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2703 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2704 template_vars
= {'url': iri_to_uri(info_dict['webpage_url'])}
2705 if link_type
== 'desktop':
2706 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2707 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
2708 except (OSError, IOError):
2709 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
2714 'url': self
.params
.get('writeurllink'),
2715 'webloc': self
.params
.get('writewebloclink'),
2716 'desktop': self
.params
.get('writedesktoplink'),
2718 if self
.params
.get('writelink'):
2719 link_type
= ('webloc' if sys
.platform
== 'darwin'
2720 else 'desktop' if sys
.platform
.startswith('linux')
2722 write_links
[link_type
] = True
2724 if any(should_write
and not _write_link_file(link_type
)
2725 for link_type
, should_write
in write_links
.items()):
2729 info_dict
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2730 except PostProcessingError
as err
:
2731 self
.report_error('Preprocessing: %s' % str(err
))
2734 must_record_download_archive
= False
2735 if self
.params
.get('skip_download', False):
2736 info_dict
['filepath'] = temp_filename
2737 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2738 info_dict
['__files_to_move'] = files_to_move
2739 info_dict
= self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
)
2742 info_dict
.setdefault('__postprocessors', [])
2745 def existing_file(*filepaths
):
2746 ext
= info_dict
.get('ext')
2747 final_ext
= self
.params
.get('final_ext', ext
)
2749 for file in orderedSet(filepaths
):
2750 if final_ext
!= ext
:
2751 converted
= replace_extension(file, final_ext
, ext
)
2752 if os
.path
.exists(encodeFilename(converted
)):
2753 existing_files
.append(converted
)
2754 if os
.path
.exists(encodeFilename(file)):
2755 existing_files
.append(file)
2757 if not existing_files
or self
.params
.get('overwrites', False):
2758 for file in orderedSet(existing_files
):
2759 self
.report_file_delete(file)
2760 os
.remove(encodeFilename(file))
2763 info_dict
['ext'] = os
.path
.splitext(existing_files
[0])[1][1:]
2764 return existing_files
[0]
2767 if info_dict
.get('requested_formats') is not None:
2769 def compatible_formats(formats
):
2770 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2771 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2772 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2773 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2777 exts
= set(format
.get('ext') for format
in formats
)
2779 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2782 for ext_sets
in COMPATIBLE_EXTS
:
2783 if ext_sets
.issuperset(exts
):
2785 # TODO: Check acodec/vcodec
2788 requested_formats
= info_dict
['requested_formats']
2789 old_ext
= info_dict
['ext']
2790 if self
.params
.get('merge_output_format') is None:
2791 if not compatible_formats(requested_formats
):
2792 info_dict
['ext'] = 'mkv'
2793 self
.report_warning(
2794 'Requested formats are incompatible for merge and will be merged into mkv')
2795 if (info_dict
['ext'] == 'webm'
2796 and info_dict
.get('thumbnails')
2797 # check with type instead of pp_key, __name__, or isinstance
2798 # since we dont want any custom PPs to trigger this
2799 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])):
2800 info_dict
['ext'] = 'mkv'
2801 self
.report_warning(
2802 'webm doesn\'t support embedding a thumbnail, mkv will be used')
2803 new_ext
= info_dict
['ext']
2805 def correct_ext(filename
, ext
=new_ext
):
2808 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2810 os
.path
.splitext(filename
)[0]
2811 if filename_real_ext
in (old_ext
, new_ext
)
2813 return '%s.%s' % (filename_wo_ext
, ext
)
2815 # Ensure filename always has a correct extension for successful merge
2816 full_filename
= correct_ext(full_filename
)
2817 temp_filename
= correct_ext(temp_filename
)
2818 dl_filename
= existing_file(full_filename
, temp_filename
)
2819 info_dict
['__real_download'] = False
2821 if dl_filename
is not None:
2822 self
.report_file_already_downloaded(dl_filename
)
2823 elif get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-'):
2824 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
2825 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2826 info_dict
['__real_download'] = real_download
2829 merger
= FFmpegMergerPP(self
)
2830 if self
.params
.get('allow_unplayable_formats'):
2831 self
.report_warning(
2832 'You have requested merging of multiple formats '
2833 'while also allowing unplayable formats to be downloaded. '
2834 'The formats won\'t be merged to prevent data corruption.')
2835 elif not merger
.available
:
2836 self
.report_warning(
2837 'You have requested merging of multiple formats but ffmpeg is not installed. '
2838 'The formats won\'t be merged.')
2840 if temp_filename
== '-':
2841 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
)
2842 else 'but the formats are incompatible for simultaneous download' if merger
.available
2843 else 'but ffmpeg is not installed')
2844 self
.report_warning(
2845 f
'You have requested downloading multiple formats to stdout {reason}. '
2846 'The formats will be streamed one after the other')
2847 fname
= temp_filename
2848 for f
in requested_formats
:
2849 new_info
= dict(info_dict
)
2850 del new_info
['requested_formats']
2852 if temp_filename
!= '-':
2853 fname
= prepend_extension(
2854 correct_ext(temp_filename
, new_info
['ext']),
2855 'f%s' % f
['format_id'], new_info
['ext'])
2856 if not self
._ensure
_dir
_exists
(fname
):
2858 f
['filepath'] = fname
2859 downloaded
.append(fname
)
2860 partial_success
, real_download
= self
.dl(fname
, new_info
)
2861 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
2862 success
= success
and partial_success
2863 if merger
.available
and not self
.params
.get('allow_unplayable_formats'):
2864 info_dict
['__postprocessors'].append(merger
)
2865 info_dict
['__files_to_merge'] = downloaded
2866 # Even if there were no downloads, it is being merged only now
2867 info_dict
['__real_download'] = True
2869 for file in downloaded
:
2870 files_to_move
[file] = None
2872 # Just a single file
2873 dl_filename
= existing_file(full_filename
, temp_filename
)
2874 if dl_filename
is None or dl_filename
== temp_filename
:
2875 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
2876 # So we should try to resume the download
2877 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2878 info_dict
['__real_download'] = real_download
2880 self
.report_file_already_downloaded(dl_filename
)
2882 dl_filename
= dl_filename
or temp_filename
2883 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2885 except network_exceptions
as err
:
2886 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
2888 except (OSError, IOError) as err
:
2889 raise UnavailableVideoError(err
)
2890 except (ContentTooShortError
, ) as err
:
2891 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
2894 if success
and full_filename
!= '-':
2898 fixup_policy
= self
.params
.get('fixup')
2899 vid
= info_dict
['id']
2901 if fixup_policy
in ('ignore', 'never'):
2903 elif fixup_policy
== 'warn':
2905 elif fixup_policy
!= 'force':
2906 assert fixup_policy
in ('detect_or_warn', None)
2907 if not info_dict
.get('__real_download'):
2910 def ffmpeg_fixup(cndn
, msg
, cls
):
2914 self
.report_warning(f
'{vid}: {msg}')
2918 info_dict
['__postprocessors'].append(pp
)
2920 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
2922 stretched_ratio
= info_dict
.get('stretched_ratio')
2924 stretched_ratio
not in (1, None),
2925 f
'Non-uniform pixel ratio {stretched_ratio}',
2926 FFmpegFixupStretchedPP
)
2929 (info_dict
.get('requested_formats') is None
2930 and info_dict
.get('container') == 'm4a_dash'
2931 and info_dict
.get('ext') == 'm4a'),
2932 'writing DASH m4a. Only some players support this container',
2935 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
2936 downloader
= downloader
.__name
__ if downloader
else None
2937 ffmpeg_fixup(info_dict
.get('requested_formats') is None and downloader
== 'HlsFD',
2938 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
2940 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
2941 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP
)
2945 info_dict
= self
.post_process(dl_filename
, info_dict
, files_to_move
)
2946 except PostProcessingError
as err
:
2947 self
.report_error('Postprocessing: %s' % str(err
))
2950 for ph
in self
._post
_hooks
:
2951 ph(info_dict
['filepath'])
2952 except Exception as err
:
2953 self
.report_error('post hooks: %s' % str(err
))
2955 must_record_download_archive
= True
2957 if must_record_download_archive
or self
.params
.get('force_write_download_archive', False):
2958 self
.record_download_archive(info_dict
)
2959 max_downloads
= self
.params
.get('max_downloads')
2960 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
2961 raise MaxDownloadsReached()
2963 def __download_wrapper(self
, func
):
2964 @functools.wraps(func
)
2965 def wrapper(*args
, **kwargs
):
2967 res
= func(*args
, **kwargs
)
2968 except UnavailableVideoError
as e
:
2969 self
.report_error(e
)
2970 except DownloadCancelled
as e
:
2971 self
.to_screen(f
'[info] {e}')
2974 if self
.params
.get('dump_single_json', False):
2975 self
.post_extract(res
)
2976 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
2979 def download(self
, url_list
):
2980 """Download a given list of URLs."""
2981 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
2982 outtmpl
= self
.outtmpl_dict
['default']
2983 if (len(url_list
) > 1
2985 and '%' not in outtmpl
2986 and self
.params
.get('max_downloads') != 1):
2987 raise SameFileError(outtmpl
)
2989 for url
in url_list
:
2990 self
.__download
_wrapper
(self
.extract_info
)(
2991 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
2993 return self
._download
_retcode
2995 def download_with_info_file(self
, info_filename
):
2996 with contextlib
.closing(fileinput
.FileInput(
2997 [info_filename
], mode
='r',
2998 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
2999 # FileInput doesn't have a read method, we can't call json.load
3000 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3002 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3003 except (DownloadError
, EntryNotInPlaylist
, ThrottledDownload
) as e
:
3004 if not isinstance(e
, EntryNotInPlaylist
):
3005 self
.to_stderr('\r')
3006 webpage_url
= info
.get('webpage_url')
3007 if webpage_url
is not None:
3008 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3009 return self
.download([webpage_url
])
3012 return self
._download
_retcode
3015 def sanitize_info(info_dict
, remove_private_keys
=False):
3016 ''' Sanitize the infodict for converting to json '''
3017 if info_dict
is None:
3019 info_dict
.setdefault('epoch', int(time
.time()))
3020 remove_keys
= {'__original_infodict'}
# Always remove this since this may contain a copy of the entire dict
3021 keep_keys
= ['_type'] # Always keep this to facilitate load-info-json
3022 if remove_private_keys
:
3024 'requested_formats', 'requested_subtitles', 'requested_entries', 'entries',
3025 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3027 empty_values
= (None, {}, [], set(), tuple())
3028 reject
= lambda k
, v
: k
not in keep_keys
and (
3029 k
.startswith('_') or k
in remove_keys
or v
in empty_values
)
3031 reject
= lambda k
, v
: k
in remove_keys
3032 filter_fn
= lambda obj
: (
3033 list(map(filter_fn
, obj
)) if isinstance(obj
, (LazyList
, list, tuple, set))
3034 else obj
if not isinstance(obj
, dict)
3035 else dict((k
, filter_fn(v
)) for k
, v
in obj
.items() if not reject(k
, v
)))
3036 return filter_fn(info_dict
)
3039 def filter_requested_info(info_dict
, actually_filter
=True):
3040 ''' Alias of sanitize_info for backward compatibility '''
3041 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3043 def run_pp(self
, pp
, infodict
):
3044 files_to_delete
= []
3045 if '__files_to_move' not in infodict
:
3046 infodict
['__files_to_move'] = {}
3048 files_to_delete
, infodict
= pp
.run(infodict
)
3049 except PostProcessingError
as e
:
3050 # Must be True and not 'only_download'
3051 if self
.params
.get('ignoreerrors') is True:
3052 self
.report_error(e
)
3056 if not files_to_delete
:
3058 if self
.params
.get('keepvideo', False):
3059 for f
in files_to_delete
:
3060 infodict
['__files_to_move'].setdefault(f
, '')
3062 for old_filename
in set(files_to_delete
):
3063 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
3065 os
.remove(encodeFilename(old_filename
))
3066 except (IOError, OSError):
3067 self
.report_warning('Unable to remove downloaded original file')
3068 if old_filename
in infodict
['__files_to_move']:
3069 del infodict
['__files_to_move'][old_filename
]
3073 def post_extract(info_dict
):
3074 def actual_post_extract(info_dict
):
3075 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3076 for video_dict
in info_dict
.get('entries', {}):
3077 actual_post_extract(video_dict
or {})
3080 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
3081 extra
= post_extractor().items()
3082 info_dict
.update(extra
)
3083 info_dict
.pop('__post_extractor', None)
3085 original_infodict
= info_dict
.get('__original_infodict') or {}
3086 original_infodict
.update(extra
)
3087 original_infodict
.pop('__post_extractor', None)
3089 actual_post_extract(info_dict
or {})
3091 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3092 info
= dict(ie_info
)
3093 info
['__files_to_move'] = files_to_move
or {}
3094 for pp
in self
._pps
[key
]:
3095 info
= self
.run_pp(pp
, info
)
3096 return info
, info
.pop('__files_to_move', None)
3098 def post_process(self
, filename
, ie_info
, files_to_move
=None):
3099 """Run all the postprocessors on the given file."""
3100 info
= dict(ie_info
)
3101 info
['filepath'] = filename
3102 info
['__files_to_move'] = files_to_move
or {}
3104 for pp
in ie_info
.get('__postprocessors', []) + self
._pps
['post_process']:
3105 info
= self
.run_pp(pp
, info
)
3106 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3107 del info
['__files_to_move']
3108 for pp
in self
._pps
['after_move']:
3109 info
= self
.run_pp(pp
, info
)
3112 def _make_archive_id(self
, info_dict
):
3113 video_id
= info_dict
.get('id')
3116 # Future-proof against any change in case
3117 # and backwards compatibility with prior versions
3118 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3119 if extractor
is None:
3120 url
= str_or_none(info_dict
.get('url'))
3123 # Try to find matching extractor for the URL and take its ie_key
3124 for ie_key
, ie
in self
._ies
.items():
3125 if ie
.suitable(url
):
3130 return '%s %s' % (extractor
.lower(), video_id
)
3132 def in_download_archive(self
, info_dict
):
3133 fn
= self
.params
.get('download_archive')
3137 vid_id
= self
._make
_archive
_id
(info_dict
)
3139 return False # Incomplete video information
3141 return vid_id
in self
.archive
3143 def record_download_archive(self
, info_dict
):
3144 fn
= self
.params
.get('download_archive')
3147 vid_id
= self
._make
_archive
_id
(info_dict
)
3149 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3150 archive_file
.write(vid_id
+ '\n')
3151 self
.archive
.add(vid_id
)
3154 def format_resolution(format
, default
='unknown'):
3155 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3157 if format
.get('resolution') is not None:
3158 return format
['resolution']
3159 if format
.get('width') and format
.get('height'):
3160 return '%dx%d' % (format
['width'], format
['height'])
3161 elif format
.get('height'):
3162 return '%sp' % format
['height']
3163 elif format
.get('width'):
3164 return '%dx?' % format
['width']
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 delim
= self
._format
_screen
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3235 self
._format
_screen
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3236 format_field(f
, 'ext'),
3237 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3238 format_field(f
, 'fps', '\t%d'),
3239 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3241 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3242 format_field(f
, 'tbr', '\t%dk'),
3243 shorten_protocol_name(f
.get('protocol', '').replace('native', 'n')),
3245 format_field(f
, 'vcodec', default
='unknown').replace(
3247 'images' if f
.get('acodec') == 'none'
3248 else self
._format
_screen
('audio only', self
.Styles
.SUPPRESS
)),
3249 format_field(f
, 'vbr', '\t%dk'),
3250 format_field(f
, 'acodec', default
='unknown').replace(
3252 '' if f
.get('vcodec') == 'none'
3253 else self
._format
_screen
('video only', self
.Styles
.SUPPRESS
)),
3254 format_field(f
, 'abr', '\t%dk'),
3255 format_field(f
, 'asr', '\t%dHz'),
3257 self
._format
_screen
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3258 format_field(f
, 'language', '[%s]'),
3260 format_field(f
, 'format_note'),
3261 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3264 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3265 header_line
= self
._list
_format
_headers
(
3266 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3267 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3271 format_field(f
, 'format_id'),
3272 format_field(f
, 'ext'),
3273 self
.format_resolution(f
),
3274 self
._format
_note
(f
)]
3276 if f
.get('preference') is None or f
['preference'] >= -1000]
3277 header_line
= ['format code', 'extension', 'resolution', 'note']
3280 '[info] Available formats for %s:' % info_dict
['id'])
3281 self
.to_stdout(render_table(
3283 extra_gap
=(0 if new_format
else 1),
3284 hide_empty
=new_format
,
3285 delim
=new_format
and self
._format
_screen
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True)))
3287 def list_thumbnails(self
, info_dict
):
3288 thumbnails
= list(info_dict
.get('thumbnails'))
3290 self
.to_screen('[info] No thumbnails present for %s' % info_dict
['id'])
3294 '[info] Thumbnails for %s:' % info_dict
['id'])
3295 self
.to_stdout(render_table(
3296 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3297 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
]))
3299 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3301 self
.to_screen('%s has no %s' % (video_id
, name
))
3304 'Available %s for %s:' % (name
, video_id
))
3306 def _row(lang
, formats
):
3307 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3308 if len(set(names
)) == 1:
3309 names
= [] if names
[0] == 'unknown' else names
[:1]
3310 return [lang
, ', '.join(names
), ', '.join(exts
)]
3312 self
.to_stdout(render_table(
3313 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3314 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3317 def urlopen(self
, req
):
3318 """ Start an HTTP download """
3319 if isinstance(req
, compat_basestring
):
3320 req
= sanitized_Request(req
)
3321 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3323 def print_debug_header(self
):
3324 if not self
.params
.get('verbose'):
3327 def get_encoding(stream
):
3328 ret
= getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__)
3329 if not supports_terminal_sequences(stream
):
3333 encoding_str
= 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3334 locale
.getpreferredencoding(),
3335 sys
.getfilesystemencoding(),
3336 get_encoding(self
._screen
_file
), get_encoding(self
._err
_file
),
3337 self
.get_encoding())
3339 logger
= self
.params
.get('logger')
3341 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3342 write_debug(encoding_str
)
3344 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3345 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3347 source
= detect_variant()
3348 write_debug('yt-dlp version %s%s' % (__version__
, '' if source
== 'unknown' else f
' ({source})'))
3349 if not _LAZY_LOADER
:
3350 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3351 write_debug('Lazy loading extractors is forcibly disabled')
3353 write_debug('Lazy loading extractors is disabled')
3354 if plugin_extractors
or plugin_postprocessors
:
3355 write_debug('Plugins: %s' % [
3356 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3357 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3358 if self
.params
.get('compat_opts'):
3359 write_debug('Compatibility options: %s' % ', '.join(self
.params
.get('compat_opts')))
3362 ['git', 'rev-parse', '--short', 'HEAD'],
3363 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3364 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3365 out
, err
= sp
.communicate_or_kill()
3366 out
= out
.decode().strip()
3367 if re
.match('[0-9a-f]+', out
):
3368 write_debug('Git HEAD: %s' % out
)
3375 def python_implementation():
3376 impl_name
= platform
.python_implementation()
3377 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3378 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3381 write_debug('Python version %s (%s %s) - %s' % (
3382 platform
.python_version(),
3383 python_implementation(),
3384 platform
.architecture()[0],
3387 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3388 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3390 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features
)
3392 exe_versions
['rtmpdump'] = rtmpdump_version()
3393 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3394 exe_str
= ', '.join(
3395 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3397 write_debug('exe versions: %s' % exe_str
)
3399 from .downloader
.websocket
import has_websockets
3400 from .postprocessor
.embedthumbnail
import has_mutagen
3401 from .cookies
import SQLITE_AVAILABLE
, KEYRING_AVAILABLE
3403 lib_str
= join_nonempty(
3404 compat_pycrypto_AES
and compat_pycrypto_AES
.__name
__.split('.')[0],
3405 KEYRING_AVAILABLE
and 'keyring',
3406 has_mutagen
and 'mutagen',
3407 SQLITE_AVAILABLE
and 'sqlite',
3408 has_websockets
and 'websockets',
3409 delim
=', ') or 'none'
3410 write_debug('Optional libraries: %s' % lib_str
)
3413 for handler
in self
._opener
.handlers
:
3414 if hasattr(handler
, 'proxies'):
3415 proxy_map
.update(handler
.proxies
)
3416 write_debug(f
'Proxy map: {proxy_map}')
3419 if False and self
.params
.get('call_home'):
3420 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3421 write_debug('Public IP address: %s' % ipaddr
)
3422 latest_version
= self
.urlopen(
3423 'https://yt-dl.org/latest/version').read().decode('utf-8')
3424 if version_tuple(latest_version
) > version_tuple(__version__
):
3425 self
.report_warning(
3426 'You are using an outdated version (newest version: %s)! '
3427 'See https://yt-dl.org/update if you need help updating.' %
3430 def _setup_opener(self
):
3431 timeout_val
= self
.params
.get('socket_timeout')
3432 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3434 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3435 opts_cookiefile
= self
.params
.get('cookiefile')
3436 opts_proxy
= self
.params
.get('proxy')
3438 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3440 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3441 if opts_proxy
is not None:
3442 if opts_proxy
== '':
3445 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3447 proxies
= compat_urllib_request
.getproxies()
3448 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3449 if 'http' in proxies
and 'https' not in proxies
:
3450 proxies
['https'] = proxies
['http']
3451 proxy_handler
= PerRequestProxyHandler(proxies
)
3453 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3454 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3455 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3456 redirect_handler
= YoutubeDLRedirectHandler()
3457 data_handler
= compat_urllib_request_DataHandler()
3459 # When passing our own FileHandler instance, build_opener won't add the
3460 # default FileHandler and allows us to disable the file protocol, which
3461 # can be used for malicious purposes (see
3462 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3463 file_handler
= compat_urllib_request
.FileHandler()
3465 def file_open(*args
, **kwargs
):
3466 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3467 file_handler
.file_open
= file_open
3469 opener
= compat_urllib_request
.build_opener(
3470 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3472 # Delete the default user-agent header, which would otherwise apply in
3473 # cases where our custom HTTP handler doesn't come into play
3474 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3475 opener
.addheaders
= []
3476 self
._opener
= opener
3478 def encode(self
, s
):
3479 if isinstance(s
, bytes):
3480 return s
# Already encoded
3483 return s
.encode(self
.get_encoding())
3484 except UnicodeEncodeError as err
:
3485 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3488 def get_encoding(self
):
3489 encoding
= self
.params
.get('encoding')
3490 if encoding
is None:
3491 encoding
= preferredencoding()
3494 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3495 ''' Write infojson and returns True = written, False = skip, None = error '''
3496 if overwrite
is None:
3497 overwrite
= self
.params
.get('overwrites', True)
3498 if not self
.params
.get('writeinfojson'):
3501 self
.write_debug(f
'Skipping writing {label} infojson')
3503 elif not self
._ensure
_dir
_exists
(infofn
):
3505 elif not overwrite
and os
.path
.exists(infofn
):
3506 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3508 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3510 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3511 except (OSError, IOError):
3512 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3516 def _write_description(self
, label
, ie_result
, descfn
):
3517 ''' Write description and returns True = written, False = skip, None = error '''
3518 if not self
.params
.get('writedescription'):
3521 self
.write_debug(f
'Skipping writing {label} description')
3523 elif not self
._ensure
_dir
_exists
(descfn
):
3525 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3526 self
.to_screen(f
'[info] {label.title()} description is already present')
3527 elif ie_result
.get('description') is None:
3528 self
.report_warning(f
'There\'s no {label} description to write')
3532 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3533 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3534 descfile
.write(ie_result
['description'])
3535 except (OSError, IOError):
3536 self
.report_error(f
'Cannot write {label} description file {descfn}')
3540 def _write_subtitles(self
, info_dict
, filename
):
3541 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3543 subtitles
= info_dict
.get('requested_subtitles')
3544 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3545 # subtitles download errors are already managed as troubles in relevant IE
3546 # that way it will silently go on when used with unsupporting IE
3549 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3550 if not sub_filename_base
:
3551 self
.to_screen('[info] Skipping writing video subtitles')
3553 for sub_lang
, sub_info
in subtitles
.items():
3554 sub_format
= sub_info
['ext']
3555 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3556 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3557 if not self
.params
.get('overwrites', True) and os
.path
.exists(sub_filename
):
3558 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3559 sub_info
['filepath'] = sub_filename
3560 ret
.append((sub_filename
, sub_filename_final
))
3563 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3564 if sub_info
.get('data') is not None:
3566 # Use newline='' to prevent conversion of newline characters
3567 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3568 with io
.open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3569 subfile
.write(sub_info
['data'])
3570 sub_info
['filepath'] = sub_filename
3571 ret
.append((sub_filename
, sub_filename_final
))
3573 except (OSError, IOError):
3574 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3578 sub_copy
= sub_info
.copy()
3579 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3580 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3581 sub_info
['filepath'] = sub_filename
3582 ret
.append((sub_filename
, sub_filename_final
))
3583 except (ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3584 self
.report_warning(f
'Unable to download video subtitles for {sub_lang!r}: {err}')
3588 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3589 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3590 write_all
= self
.params
.get('write_all_thumbnails', False)
3591 thumbnails
, ret
= [], []
3592 if write_all
or self
.params
.get('writethumbnail', False):
3593 thumbnails
= info_dict
.get('thumbnails') or []
3594 multiple
= write_all
and len(thumbnails
) > 1
3596 if thumb_filename_base
is None:
3597 thumb_filename_base
= filename
3598 if thumbnails
and not thumb_filename_base
:
3599 self
.write_debug(f
'Skipping writing {label} thumbnail')
3602 for t
in thumbnails
[::-1]:
3603 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3604 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3605 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3606 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3608 if not self
.params
.get('overwrites', True) and os
.path
.exists(thumb_filename
):
3609 ret
.append((thumb_filename
, thumb_filename_final
))
3610 t
['filepath'] = thumb_filename
3611 self
.to_screen('[info] %s is already present' % (
3612 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3614 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3616 uf
= self
.urlopen(t
['url'])
3617 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3618 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3619 shutil
.copyfileobj(uf
, thumbf
)
3620 ret
.append((thumb_filename
, thumb_filename_final
))
3621 t
['filepath'] = thumb_filename
3622 except network_exceptions
as err
:
3623 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3624 if ret
and not write_all
: