27 from string
import ascii_letters
29 from .cache
import Cache
32 compat_get_terminal_size
,
38 compat_urllib_request
,
39 windows_enable_vt_mode
,
41 from .cookies
import load_cookies
42 from .downloader
import FFmpegFD
, get_suitable_downloader
, shorten_protocol_name
43 from .downloader
.rtmp
import rtmpdump_version
44 from .extractor
import _LAZY_LOADER
45 from .extractor
import _PLUGIN_CLASSES
as plugin_extractors
46 from .extractor
import gen_extractor_classes
, get_info_extractor
47 from .extractor
.openload
import PhantomJSwrapper
48 from .minicurses
import format_text
49 from .postprocessor
import _PLUGIN_CLASSES
as plugin_postprocessors
50 from .postprocessor
import (
52 FFmpegFixupDuplicateMoovPP
,
53 FFmpegFixupDurationPP
,
56 FFmpegFixupStretchedPP
,
57 FFmpegFixupTimestampPP
,
60 MoveFilesAfterDownloadPP
,
63 from .update
import detect_variant
86 PerRequestProxyHandler
,
92 UnavailableVideoError
,
93 YoutubeDLCookieProcessor
,
95 YoutubeDLRedirectHandler
,
108 format_decimal_suffix
,
127 register_socks_protocols
,
128 remove_terminal_sequences
,
139 supports_terminal_sequences
,
150 from .version
import RELEASE_GIT_HEAD
, __version__
152 if compat_os_name
== 'nt':
159 YoutubeDL objects are the ones responsible of downloading the
160 actual video file and writing it to disk if the user has requested
161 it, among some other tasks. In most cases there should be one per
162 program. As, given a video URL, the downloader doesn't know how to
163 extract all the needed information, task that InfoExtractors do, it
164 has to pass the URL to one of them.
166 For this, YoutubeDL objects have a method that allows
167 InfoExtractors to be registered in a given order. When it is passed
168 a URL, the YoutubeDL object handles it to the first InfoExtractor it
169 finds that reports being able to handle it. The InfoExtractor extracts
170 all the information about the video or videos the URL refers to, and
171 YoutubeDL process the extracted information, possibly using a File
172 Downloader to download the video.
174 YoutubeDL objects accept a lot of parameters. In order not to saturate
175 the object constructor with arguments, it receives a dictionary of
176 options instead. These options are available through the params
177 attribute for the InfoExtractors to use. The YoutubeDL also
178 registers itself as the downloader in charge for the InfoExtractors
179 that are added to it, so this is a "mutual registration".
183 username: Username for authentication purposes.
184 password: Password for authentication purposes.
185 videopassword: Password for accessing a video.
186 ap_mso: Adobe Pass multiple-system operator identifier.
187 ap_username: Multiple-system operator account username.
188 ap_password: Multiple-system operator account password.
189 usenetrc: Use netrc for authentication instead.
190 verbose: Print additional info to stdout.
191 quiet: Do not print messages to stdout.
192 no_warnings: Do not print out anything for warnings.
193 forceprint: A dict with keys WHEN mapped to a list of templates to
194 print to stdout. The allowed keys are video or any of the
195 items in utils.POSTPROCESS_WHEN.
196 For compatibility, a single list is also accepted
197 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
198 a list of tuples with (template, filename)
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 prefer_free_formats: Whether to prefer video formats with free containers
226 over non-free ones of same quality.
227 allow_multiple_video_streams: Allow multiple video streams to be merged
229 allow_multiple_audio_streams: Allow multiple audio streams to be merged
231 check_formats Whether to test if the formats are downloadable.
232 Can be True (check all), False (check none),
233 'selected' (check selected formats),
234 or None (check only if requested by extractor)
235 paths: Dictionary of output paths. The allowed keys are 'home'
236 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
237 outtmpl: Dictionary of templates for output names. Allowed keys
238 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
239 For compatibility with youtube-dl, a single string can also be used
240 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
241 restrictfilenames: Do not allow "&" and spaces in file names
242 trim_file_name: Limit length of filename (extension excluded)
243 windowsfilenames: Force the filenames to be windows compatible
244 ignoreerrors: Do not stop on download/postprocessing errors.
245 Can be 'only_download' to ignore only download errors.
246 Default is 'only_download' for CLI, but False for API
247 skip_playlist_after_errors: Number of allowed failures until the rest of
248 the playlist is skipped
249 force_generic_extractor: Force downloader to use the generic extractor
250 overwrites: Overwrite all video and metadata files if True,
251 overwrite only non-video files if None
252 and don't overwrite any file if False
253 For compatibility with youtube-dl,
254 "nooverwrites" may also be used instead
255 playliststart: Playlist item to start at.
256 playlistend: Playlist item to end at.
257 playlist_items: Specific indices of playlist to download.
258 playlistreverse: Download playlist items in reverse order.
259 playlistrandom: Download playlist items in random order.
260 matchtitle: Download only matching titles.
261 rejecttitle: Reject downloads for matching titles.
262 logger: Log messages to a logging.Logger instance.
263 logtostderr: Log messages to stderr instead of stdout.
264 consoletitle: Display progress in console window's titlebar.
265 writedescription: Write the video description to a .description file
266 writeinfojson: Write the video description to a .info.json file
267 clean_infojson: Remove private fields from the infojson
268 getcomments: Extract video comments. This will not be written to disk
269 unless writeinfojson is also given
270 writeannotations: Write the video annotations to a .annotations.xml file
271 writethumbnail: Write the thumbnail image to a file
272 allow_playlist_files: Whether to write playlists' description, infojson etc
273 also to disk when using the 'write*' options
274 write_all_thumbnails: Write all thumbnail formats to files
275 writelink: Write an internet shortcut file, depending on the
276 current platform (.url/.webloc/.desktop)
277 writeurllink: Write a Windows internet shortcut file (.url)
278 writewebloclink: Write a macOS internet shortcut file (.webloc)
279 writedesktoplink: Write a Linux internet shortcut file (.desktop)
280 writesubtitles: Write the video subtitles to a file
281 writeautomaticsub: Write the automatically generated subtitles to a file
282 allsubtitles: Deprecated - Use subtitleslangs = ['all']
283 Downloads all the subtitles of the video
284 (requires writesubtitles or writeautomaticsub)
285 listsubtitles: Lists all available subtitles for the video
286 subtitlesformat: The format code for subtitles
287 subtitleslangs: List of languages of the subtitles to download (can be regex).
288 The list may contain "all" to refer to all the available
289 subtitles. The language can be prefixed with a "-" to
290 exclude it from the requested languages. Eg: ['all', '-live_chat']
291 keepvideo: Keep the video file after post-processing
292 daterange: A DateRange object, download only if the upload_date is in the range.
293 skip_download: Skip the actual download of the video file
294 cachedir: Location of the cache files in the filesystem.
295 False to disable filesystem cache.
296 noplaylist: Download single video instead of a playlist if in doubt.
297 age_limit: An integer representing the user's age in years.
298 Unsuitable videos for the given age are skipped.
299 min_views: An integer representing the minimum view count the video
300 must have in order to not be skipped.
301 Videos without view count information are always
302 downloaded. None for no limit.
303 max_views: An integer representing the maximum view count.
304 Videos that are more popular than that are not
306 Videos without view count information are always
307 downloaded. None for no limit.
308 download_archive: File name of a file where all downloads are recorded.
309 Videos already present in the file are not downloaded
311 break_on_existing: Stop the download process after attempting to download a
312 file that is in the archive.
313 break_on_reject: Stop the download process when encountering a video that
314 has been filtered out.
315 break_per_url: Whether break_on_reject and break_on_existing
316 should act on each input URL as opposed to for the entire queue
317 cookiefile: File name where cookies should be read from and dumped to
318 cookiesfrombrowser: A tuple containing the name of the browser, the profile
319 name/pathfrom where cookies are loaded, and the name of the
320 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
321 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
322 support RFC 5746 secure renegotiation
323 nocheckcertificate: Do not verify SSL certificates
324 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
325 At the moment, this is only supported by YouTube.
326 http_headers: A dictionary of custom headers to be used for all requests
327 proxy: URL of the proxy server to use
328 geo_verification_proxy: URL of the proxy to use for IP address verification
329 on geo-restricted sites.
330 socket_timeout: Time to wait for unresponsive hosts, in seconds
331 bidi_workaround: Work around buggy terminals without bidirectional text
332 support, using fridibi
333 debug_printtraffic:Print out sent and received HTTP traffic
334 include_ads: Download ads as well (deprecated)
335 default_search: Prepend this string if an input url is not valid.
336 'auto' for elaborate guessing
337 encoding: Use this encoding instead of the system-specified.
338 extract_flat: Do not resolve URLs, return the immediate result.
339 Pass in 'in_playlist' to only show this behavior for
341 wait_for_video: If given, wait for scheduled streams to become available.
342 The value should be a tuple containing the range
343 (min_secs, max_secs) to wait between retries
344 postprocessors: A list of dictionaries, each with an entry
345 * key: The name of the postprocessor. See
346 yt_dlp/postprocessor/__init__.py for a list.
347 * when: When to run the postprocessor. Allowed values are
348 the entries of utils.POSTPROCESS_WHEN
349 Assumed to be 'post_process' if not given
350 post_hooks: Deprecated - Register a custom postprocessor instead
351 A list of functions that get called as the final step
352 for each video file, after all postprocessors have been
353 called. The filename will be passed as the only argument.
354 progress_hooks: A list of functions that get called on download
355 progress, with a dictionary with the entries
356 * status: One of "downloading", "error", or "finished".
357 Check this first and ignore unknown values.
358 * info_dict: The extracted info_dict
360 If status is one of "downloading", or "finished", the
361 following properties may also be present:
362 * filename: The final filename (always present)
363 * tmpfilename: The filename we're currently writing to
364 * downloaded_bytes: Bytes on disk
365 * total_bytes: Size of the whole file, None if unknown
366 * total_bytes_estimate: Guess of the eventual file size,
368 * elapsed: The number of seconds since download started.
369 * eta: The estimated time in seconds, None if unknown
370 * speed: The download speed in bytes/second, None if
372 * fragment_index: The counter of the currently
373 downloaded video fragment.
374 * fragment_count: The number of fragments (= individual
375 files that will be merged)
377 Progress hooks are guaranteed to be called at least once
378 (with status "finished") if the download is successful.
379 postprocessor_hooks: A list of functions that get called on postprocessing
380 progress, with a dictionary with the entries
381 * status: One of "started", "processing", or "finished".
382 Check this first and ignore unknown values.
383 * postprocessor: Name of the postprocessor
384 * info_dict: The extracted info_dict
386 Progress hooks are guaranteed to be called at least twice
387 (with status "started" and "finished") if the processing is successful.
388 merge_output_format: Extension to use when merging formats.
389 final_ext: Expected final extension; used to detect when the file was
390 already downloaded and converted
391 fixup: Automatically correct known faults of the file.
393 - "never": do nothing
394 - "warn": only emit a warning
395 - "detect_or_warn": check whether we can do anything
396 about it, warn otherwise (default)
397 source_address: Client-side IP address to bind to.
398 call_home: Boolean, true iff we are allowed to contact the
399 yt-dlp servers for debugging. (BROKEN)
400 sleep_interval_requests: Number of seconds to sleep between requests
402 sleep_interval: Number of seconds to sleep before each download when
403 used alone or a lower bound of a range for randomized
404 sleep before each download (minimum possible number
405 of seconds to sleep) when used along with
407 max_sleep_interval:Upper bound of a range for randomized sleep before each
408 download (maximum possible number of seconds to sleep).
409 Must only be used along with sleep_interval.
410 Actual sleep time will be a random float from range
411 [sleep_interval; max_sleep_interval].
412 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
413 listformats: Print an overview of available video formats and exit.
414 list_thumbnails: Print a table of all thumbnails and exit.
415 match_filter: A function that gets called with the info_dict of
417 If it returns a message, the video is ignored.
418 If it returns None, the video is downloaded.
419 match_filter_func in utils.py is one example for this.
420 no_color: Do not emit color codes in output.
421 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
424 Two-letter ISO 3166-2 country code that will be used for
425 explicit geographic restriction bypassing via faking
426 X-Forwarded-For HTTP header
428 IP range in CIDR notation that will be used similarly to
431 The following options determine which downloader is picked:
432 external_downloader: A dictionary of protocol keys and the executable of the
433 external downloader to use for it. The allowed protocols
434 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
435 Set the value to 'native' to use the native downloader
436 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
437 or {'m3u8': 'ffmpeg'} instead.
438 Use the native HLS downloader instead of ffmpeg/avconv
439 if True, otherwise use ffmpeg/avconv if False, otherwise
440 use downloader suggested by extractor if None.
441 compat_opts: Compatibility options. See "Differences in default behavior".
442 The following options do not work when used through the API:
443 filename, abort-on-error, multistreams, no-live-chat, format-sort
444 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
445 Refer __init__.py for their implementation
446 progress_template: Dictionary of templates for progress outputs.
447 Allowed keys are 'download', 'postprocess',
448 'download-title' (console title) and 'postprocess-title'.
449 The template is mapped on a dictionary with keys 'progress' and 'info'
451 The following parameters are not used by YoutubeDL itself, they are used by
452 the downloader (see yt_dlp/downloader/common.py):
453 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
454 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
455 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
456 external_downloader_args, concurrent_fragment_downloads.
458 The following options are used by the post processors:
459 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
460 otherwise prefer ffmpeg. (avconv support is deprecated)
461 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
462 to the binary or its containing directory.
463 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
464 and a list of additional command-line arguments for the
465 postprocessor/executable. The dict can also have "PP+EXE" keys
466 which are used when the given exe is used by the given PP.
467 Use 'default' as the name for arguments to passed to all PP
468 For compatibility with youtube-dl, a single list of args
471 The following options are used by the extractors:
472 extractor_retries: Number of times to retry for known errors
473 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
474 hls_split_discontinuity: Split HLS playlists to different formats at
475 discontinuities such as ad breaks (default: False)
476 extractor_args: A dictionary of arguments to be passed to the extractors.
477 See "EXTRACTOR ARGUMENTS" for details.
478 Eg: {'youtube': {'skip': ['dash', 'hls']}}
479 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
480 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
481 If True (default), DASH manifests and related
482 data will be downloaded and processed by extractor.
483 You can reduce network I/O by disabling it if you don't
484 care about DASH. (only for youtube)
485 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
486 If True (default), HLS manifests and related
487 data will be downloaded and processed by extractor.
488 You can reduce network I/O by disabling it if you don't
489 care about HLS. (only for youtube)
493 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
494 'timestamp', 'release_timestamp',
495 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
496 'average_rating', 'comment_count', 'age_limit',
497 'start_time', 'end_time',
498 'chapter_number', 'season_number', 'episode_number',
499 'track_number', 'disc_number', 'release_year',
503 # NB: Keep in sync with the docstring of extractor/common.py
504 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
505 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
506 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
507 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
508 'preference', 'language', 'language_preference', 'quality', 'source_preference',
509 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
510 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
512 _format_selection_exts
= {
513 'audio': {'m4a', 'mp3', 'ogg', 'aac'}
,
514 'video': {'mp4', 'flv', 'webm', '3gp'}
,
515 'storyboards': {'mhtml'}
,
518 def __init__(self
, params
=None, auto_init
=True):
519 """Create a FileDownloader object with the given options.
520 @param auto_init Whether to load the default extractors and print header (if verbose).
521 Set to 'no_verbose_header' to not print the header
527 self
._ies
_instances
= {}
528 self
._pps
= {k: [] for k in POSTPROCESS_WHEN}
529 self
._printed
_messages
= set()
530 self
._first
_webpage
_request
= True
531 self
._post
_hooks
= []
532 self
._progress
_hooks
= []
533 self
._postprocessor
_hooks
= []
534 self
._download
_retcode
= 0
535 self
._num
_downloads
= 0
537 self
._playlist
_level
= 0
538 self
._playlist
_urls
= set()
539 self
.cache
= Cache(self
)
541 windows_enable_vt_mode()
544 'print': sys
.stderr
if self
.params
.get('logtostderr') else sys
.stdout
,
545 'console': None if compat_os_name
== 'nt' else next(
546 filter(supports_terminal_sequences
, (sys
.stderr
, sys
.stdout
)), None)
548 self
._out
_files
['screen'] = sys
.stderr
if self
.params
.get('quiet') else self
._out
_files
['print']
549 self
._allow
_colors
= {
550 type_
: not self
.params
.get('no_color') and supports_terminal_sequences(self
._out
_files
[type_
])
551 for type_
in ('screen', 'error')
554 if sys
.version_info
< (3, 6):
556 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
558 if self
.params
.get('allow_unplayable_formats'):
560 f
'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
561 'This is a developer option intended for debugging. \n'
562 ' If you experience any issues while using this option, '
563 f
'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
565 def check_deprecated(param
, option
, suggestion
):
566 if self
.params
.get(param
) is not None:
567 self
.report_warning(f
'{option} is deprecated. Use {suggestion} instead')
571 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
572 if self
.params
.get('geo_verification_proxy') is None:
573 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
575 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
576 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
577 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
579 for msg
in self
.params
.get('_warnings', []):
580 self
.report_warning(msg
)
581 for msg
in self
.params
.get('_deprecation_warnings', []):
582 self
.deprecation_warning(msg
)
584 if 'list-formats' in self
.params
.get('compat_opts', []):
585 self
.params
['listformats_table'] = False
587 if 'overwrites' not in self
.params
and self
.params
.get('nooverwrites') is not None:
588 # nooverwrites was unnecessarily changed to overwrites
589 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
590 # This ensures compatibility with both keys
591 self
.params
['overwrites'] = not self
.params
['nooverwrites']
592 elif self
.params
.get('overwrites') is None:
593 self
.params
.pop('overwrites', None)
595 self
.params
['nooverwrites'] = not self
.params
['overwrites']
597 self
.params
.setdefault('forceprint', {})
598 self
.params
.setdefault('print_to_file', {})
600 # Compatibility with older syntax
601 if not isinstance(params
['forceprint'], dict):
602 self
.params
['forceprint'] = {'video': params['forceprint']}
604 if self
.params
.get('bidi_workaround', False):
607 master
, slave
= pty
.openpty()
608 width
= compat_get_terminal_size().columns
612 width_args
= ['-w', str(width
)]
614 stdin
=subprocess
.PIPE
,
616 stderr
=self
._out
_files
['error'])
618 self
._output
_process
= Popen(['bidiv'] + width_args
, **sp_kwargs
)
620 self
._output
_process
= Popen(['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
621 self
._output
_channel
= os
.fdopen(master
, 'rb')
622 except OSError as ose
:
623 if ose
.errno
== errno
.ENOENT
:
625 'Could not find fribidi executable, ignoring --bidi-workaround. '
626 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
631 if auto_init
!= 'no_verbose_header':
632 self
.print_debug_header()
633 self
.add_default_info_extractors()
635 if (sys
.platform
!= 'win32'
636 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
637 and not self
.params
.get('restrictfilenames', False)):
638 # Unicode filesystem API will throw errors (#1474, #13027)
640 'Assuming --restrict-filenames since file system encoding '
641 'cannot encode all characters. '
642 'Set the LC_ALL environment variable to fix this.')
643 self
.params
['restrictfilenames'] = True
645 self
.outtmpl_dict
= self
.parse_outtmpl()
647 # Creating format selector here allows us to catch syntax errors before the extraction
648 self
.format_selector
= (
649 self
.params
.get('format') if self
.params
.get('format') in (None, '-')
650 else self
.params
['format'] if callable(self
.params
['format'])
651 else self
.build_format_selector(self
.params
['format']))
653 # Set http_headers defaults according to std_headers
654 self
.params
['http_headers'] = merge_headers(std_headers
, self
.params
.get('http_headers', {}))
657 'post_hooks': self
.add_post_hook
,
658 'progress_hooks': self
.add_progress_hook
,
659 'postprocessor_hooks': self
.add_postprocessor_hook
,
661 for opt
, fn
in hooks
.items():
662 for ph
in self
.params
.get(opt
, []):
665 for pp_def_raw
in self
.params
.get('postprocessors', []):
666 pp_def
= dict(pp_def_raw
)
667 when
= pp_def
.pop('when', 'post_process')
668 self
.add_post_processor(
669 get_postprocessor(pp_def
.pop('key'))(self
, **pp_def
),
673 register_socks_protocols()
675 def preload_download_archive(fn
):
676 """Preload the archive, if any is specified"""
679 self
.write_debug(f
'Loading archive file {fn!r}')
681 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
682 for line
in archive_file
:
683 self
.archive
.add(line
.strip())
684 except OSError as ioe
:
685 if ioe
.errno
!= errno
.ENOENT
:
691 preload_download_archive(self
.params
.get('download_archive'))
693 def warn_if_short_id(self
, argv
):
694 # short YouTube ID starting with dash?
696 i
for i
, a
in enumerate(argv
)
697 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
701 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
702 + ['--'] + [argv
[i
] for i
in idxs
]
705 'Long argument string detected. '
706 'Use -- to separate parameters and URLs, like this:\n%s' %
707 args_to_str(correct_argv
))
709 def add_info_extractor(self
, ie
):
710 """Add an InfoExtractor object to the end of the list."""
712 self
._ies
[ie_key
] = ie
713 if not isinstance(ie
, type):
714 self
._ies
_instances
[ie_key
] = ie
715 ie
.set_downloader(self
)
717 def _get_info_extractor_class(self
, ie_key
):
718 ie
= self
._ies
.get(ie_key
)
720 ie
= get_info_extractor(ie_key
)
721 self
.add_info_extractor(ie
)
724 def get_info_extractor(self
, ie_key
):
726 Get an instance of an IE with name ie_key, it will try to get one from
727 the _ies list, if there's no instance it will create a new one and add
728 it to the extractor list.
730 ie
= self
._ies
_instances
.get(ie_key
)
732 ie
= get_info_extractor(ie_key
)()
733 self
.add_info_extractor(ie
)
736 def add_default_info_extractors(self
):
738 Add the InfoExtractors returned by gen_extractors to the end of the list
740 for ie
in gen_extractor_classes():
741 self
.add_info_extractor(ie
)
743 def add_post_processor(self
, pp
, when
='post_process'):
744 """Add a PostProcessor object to the end of the chain."""
745 self
._pps
[when
].append(pp
)
746 pp
.set_downloader(self
)
748 def add_post_hook(self
, ph
):
749 """Add the post hook"""
750 self
._post
_hooks
.append(ph
)
752 def add_progress_hook(self
, ph
):
753 """Add the download progress hook"""
754 self
._progress
_hooks
.append(ph
)
756 def add_postprocessor_hook(self
, ph
):
757 """Add the postprocessing progress hook"""
758 self
._postprocessor
_hooks
.append(ph
)
759 for pps
in self
._pps
.values():
761 pp
.add_progress_hook(ph
)
763 def _bidi_workaround(self
, message
):
764 if not hasattr(self
, '_output_channel'):
767 assert hasattr(self
, '_output_process')
768 assert isinstance(message
, compat_str
)
769 line_count
= message
.count('\n') + 1
770 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
771 self
._output
_process
.stdin
.flush()
772 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
773 for _
in range(line_count
))
774 return res
[:-len('\n')]
776 def _write_string(self
, message
, out
=None, only_once
=False):
778 if message
in self
._printed
_messages
:
780 self
._printed
_messages
.add(message
)
781 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
783 def to_stdout(self
, message
, skip_eol
=False, quiet
=None):
784 """Print message to stdout"""
785 if quiet
is not None:
786 self
.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
788 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
789 self
._out
_files
['print'])
791 def to_screen(self
, message
, skip_eol
=False, quiet
=None):
792 """Print message to screen if not in quiet mode"""
793 if self
.params
.get('logger'):
794 self
.params
['logger'].debug(message
)
796 if (self
.params
.get('quiet') if quiet
is None else quiet
) and not self
.params
.get('verbose'):
799 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
800 self
._out
_files
['screen'])
802 def to_stderr(self
, message
, only_once
=False):
803 """Print message to stderr"""
804 assert isinstance(message
, compat_str
)
805 if self
.params
.get('logger'):
806 self
.params
['logger'].error(message
)
808 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._out
_files
['error'], only_once
=only_once
)
810 def _send_console_code(self
, code
):
811 if compat_os_name
== 'nt' or not self
._out
_files
['console']:
813 self
._write
_string
(code
, self
._out
_files
['console'])
815 def to_console_title(self
, message
):
816 if not self
.params
.get('consoletitle', False):
818 message
= remove_terminal_sequences(message
)
819 if compat_os_name
== 'nt':
820 if ctypes
.windll
.kernel32
.GetConsoleWindow():
821 # c_wchar_p() might not be necessary if `message` is
822 # already of type unicode()
823 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
825 self
._send
_console
_code
(f
'\033]0;{message}\007')
827 def save_console_title(self
):
828 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
830 self
._send
_console
_code
('\033[22;0t') # Save the title on stack
832 def restore_console_title(self
):
833 if not self
.params
.get('consoletitle') or self
.params
.get('simulate'):
835 self
._send
_console
_code
('\033[23;0t') # Restore the title from stack
838 self
.save_console_title()
841 def __exit__(self
, *args
):
842 self
.restore_console_title()
844 if self
.params
.get('cookiefile') is not None:
845 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
847 def trouble(self
, message
=None, tb
=None, is_error
=True):
848 """Determine action to take when a download problem appears.
850 Depending on if the downloader has been configured to ignore
851 download errors or not, this method may throw an exception or
852 not when errors are found, after printing the message.
854 @param tb If given, is additional traceback information
855 @param is_error Whether to raise error according to ignorerrors
857 if message
is not None:
858 self
.to_stderr(message
)
859 if self
.params
.get('verbose'):
861 if sys
.exc_info()[0]: # if .trouble has been called from an except block
863 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
864 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
865 tb
+= encode_compat_str(traceback
.format_exc())
867 tb_data
= traceback
.format_list(traceback
.extract_stack())
868 tb
= ''.join(tb_data
)
873 if not self
.params
.get('ignoreerrors'):
874 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
875 exc_info
= sys
.exc_info()[1].exc_info
877 exc_info
= sys
.exc_info()
878 raise DownloadError(message
, exc_info
)
879 self
._download
_retcode
= 1
883 EMPHASIS
= 'light blue'
888 SUPPRESS
= 'light black'
890 def _format_text(self
, handle
, allow_colors
, text
, f
, fallback
=None, *, test_encoding
=False):
894 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
895 encoding
= self
.params
.get('encoding') or getattr(handle
, 'encoding', None) or 'ascii'
896 text
= text
.encode(encoding
, 'ignore').decode(encoding
)
897 if fallback
is not None and text
!= original_text
:
899 if isinstance(f
, Enum
):
901 return format_text(text
, f
) if allow_colors
else text
if fallback
is None else fallback
903 def _format_screen(self
, *args
, **kwargs
):
904 return self
._format
_text
(
905 self
._out
_files
['screen'], self
._allow
_colors
['screen'], *args
, **kwargs
)
907 def _format_err(self
, *args
, **kwargs
):
908 return self
._format
_text
(
909 self
._out
_files
['error'], self
._allow
_colors
['error'], *args
, **kwargs
)
911 def report_warning(self
, message
, only_once
=False):
913 Print the message to stderr, it will be prefixed with 'WARNING:'
914 If stderr is a tty file the 'WARNING:' will be colored
916 if self
.params
.get('logger') is not None:
917 self
.params
['logger'].warning(message
)
919 if self
.params
.get('no_warnings'):
921 self
.to_stderr(f
'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once
)
923 def deprecation_warning(self
, message
):
924 if self
.params
.get('logger') is not None:
925 self
.params
['logger'].warning(f
'DeprecationWarning: {message}')
927 self
.to_stderr(f
'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
929 def report_error(self
, message
, *args
, **kwargs
):
931 Do the same as trouble, but prefixes the message with 'ERROR:', colored
932 in red if stderr is a tty file.
934 self
.trouble(f
'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args
, **kwargs
)
936 def write_debug(self
, message
, only_once
=False):
937 '''Log debug message or Print message to stderr'''
938 if not self
.params
.get('verbose', False):
940 message
= '[debug] %s' % message
941 if self
.params
.get('logger'):
942 self
.params
['logger'].debug(message
)
944 self
.to_stderr(message
, only_once
)
946 def report_file_already_downloaded(self
, file_name
):
947 """Report file has already been fully downloaded."""
949 self
.to_screen('[download] %s has already been downloaded' % file_name
)
950 except UnicodeEncodeError:
951 self
.to_screen('[download] The file has already been downloaded')
953 def report_file_delete(self
, file_name
):
954 """Report that existing file will be deleted."""
956 self
.to_screen('Deleting existing file %s' % file_name
)
957 except UnicodeEncodeError:
958 self
.to_screen('Deleting existing file')
960 def raise_no_formats(self
, info
, forced
=False, *, msg
=None):
961 has_drm
= info
.get('__has_drm')
962 ignored
, expected
= self
.params
.get('ignore_no_formats_error'), bool(msg
)
963 msg
= msg
or has_drm
and 'This video is DRM protected' or 'No video formats found!'
964 if forced
or not ignored
:
965 raise ExtractorError(msg
, video_id
=info
['id'], ie
=info
['extractor'],
966 expected
=has_drm
or ignored
or expected
)
968 self
.report_warning(msg
)
970 def parse_outtmpl(self
):
971 outtmpl_dict
= self
.params
.get('outtmpl', {})
972 if not isinstance(outtmpl_dict
, dict):
973 outtmpl_dict
= {'default': outtmpl_dict}
974 # Remove spaces in the default template
975 if self
.params
.get('restrictfilenames'):
976 sanitize
= lambda x
: x
.replace(' - ', ' ').replace(' ', '-')
978 sanitize
= lambda x
: x
979 outtmpl_dict
.update({
980 k
: sanitize(v
) for k
, v
in DEFAULT_OUTTMPL
.items()
981 if outtmpl_dict
.get(k
) is None})
982 for _
, val
in outtmpl_dict
.items():
983 if isinstance(val
, bytes):
984 self
.report_warning('Parameter outtmpl is bytes, but should be a unicode string')
987 def get_output_path(self
, dir_type
='', filename
=None):
988 paths
= self
.params
.get('paths', {})
989 assert isinstance(paths
, dict)
991 expand_path(paths
.get('home', '').strip()),
992 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
994 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
997 def _outtmpl_expandpath(outtmpl
):
998 # expand_path translates '%%' into '%' and '$$' into '$'
999 # correspondingly that is not what we want since we need to keep
1000 # '%%' intact for template dict substitution step. Working around
1001 # with boundary-alike separator hack.
1002 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
1003 outtmpl
= outtmpl
.replace('%%', f
'%{sep}%').replace('$$', f
'${sep}$')
1005 # outtmpl should be expand_path'ed before template dict substitution
1006 # because meta fields may contain env variables we don't want to
1007 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1008 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1009 return expand_path(outtmpl
).replace(sep
, '')
1012 def escape_outtmpl(outtmpl
):
1013 ''' Escape any remaining strings like %s, %abc% etc. '''
1015 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
1016 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
1020 def validate_outtmpl(cls
, outtmpl
):
1021 ''' @return None or Exception object '''
1023 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljqBUDS]'),
1024 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
1025 cls
._outtmpl
_expandpath
(outtmpl
))
1027 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
1029 except ValueError as err
:
1033 def _copy_infodict(info_dict
):
1034 info_dict
= dict(info_dict
)
1035 info_dict
.pop('__postprocessors', None)
1038 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=False):
1039 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1040 @param sanitize Whether to sanitize the output as a filename.
1041 For backward compatibility, a function can also be passed
1044 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
1046 info_dict
= self
._copy
_infodict
(info_dict
)
1047 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1048 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
1049 if info_dict
.get('duration', None) is not None
1051 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
1052 info_dict
['video_autonumber'] = self
._num
_videos
1053 if info_dict
.get('resolution') is None:
1054 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
1056 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1057 # of %(field)s to %(field)0Nd for backward compatibility
1058 field_size_compat_map
= {
1059 'playlist_index': number_of_digits(info_dict
.get('_last_playlist_index') or 0),
1060 'playlist_autonumber': number_of_digits(info_dict
.get('n_entries') or 0),
1061 'autonumber': self
.params
.get('autonumber_size') or 5,
1065 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljqBUDS]'))
1070 # Field is of the form key1.key2...
1071 # where keys (except first) can be string, int or slice
1072 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
1073 MATH_FIELD_RE
= r
'''(?:{field}|{num})'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
1074 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
1075 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
1078 (?P<maths>(?:{math_op}{math_field})*)
1079 (?:>(?P<strf_format>.+?))?
1081 (?P<alternate>(?<!\\),[^|&)]+)?
1082 (?:&(?P<replacement>.*?))?
1083 (?:\|(?P<default>.*?))?
1084 )$'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
1086 def _traverse_infodict(k
):
1090 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
1092 def get_value(mdict
):
1094 value
= _traverse_infodict(mdict
['fields'])
1097 value
= float_or_none(value
)
1098 if value
is not None:
1101 offset_key
= mdict
['maths']
1103 value
= float_or_none(value
)
1107 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
1108 offset_key
).group(0)
1109 offset_key
= offset_key
[len(item
):]
1110 if operator
is None:
1111 operator
= MATH_FUNCTIONS
[item
]
1113 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
1114 offset
= float_or_none(item
)
1116 offset
= float_or_none(_traverse_infodict(item
))
1118 value
= operator(value
, multiplier
* offset
)
1119 except (TypeError, ZeroDivisionError):
1122 # Datetime formatting
1123 if mdict
['strf_format']:
1124 value
= strftime_or_none(value
, mdict
['strf_format'].replace('\\,', ','))
1128 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
1130 def filename_sanitizer(key
, value
, restricted
=self
.params
.get('restrictfilenames')):
1131 return sanitize_filename(str(value
), restricted
=restricted
, is_id
=(
1132 bool(re
.search(r
'(^|[_.])id(\.|$)', key
))
1133 if 'filename-sanitization' in self
.params
.get('compat_opts', [])
1136 sanitizer
= sanitize
if callable(sanitize
) else filename_sanitizer
1137 sanitize
= bool(sanitize
)
1139 def _dumpjson_default(obj
):
1140 if isinstance(obj
, (set, LazyList
)):
1144 def create_key(outer_mobj
):
1145 if not outer_mobj
.group('has_key'):
1146 return outer_mobj
.group(0)
1147 key
= outer_mobj
.group('key')
1148 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1149 initial_field
= mobj
.group('fields') if mobj
else ''
1150 value
, replacement
, default
= None, None, na
1152 mobj
= mobj
.groupdict()
1153 default
= mobj
['default'] if mobj
['default'] is not None else default
1154 value
= get_value(mobj
)
1155 replacement
= mobj
['replacement']
1156 if value
is None and mobj
['alternate']:
1157 mobj
= re
.match(INTERNAL_FORMAT_RE
, mobj
['remaining'][1:])
1161 fmt
= outer_mobj
.group('format')
1162 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1163 fmt
= f
'0{field_size_compat_map[key]:d}d'
1165 value
= default
if value
is None else value
if replacement
is None else replacement
1167 flags
= outer_mobj
.group('conversion') or ''
1168 str_fmt
= f
'{fmt[:-1]}s'
1169 if fmt
[-1] == 'l': # list
1170 delim
= '\n' if '#' in flags
else ', '
1171 value
, fmt
= delim
.join(map(str, variadic(value
, allowed_types
=(str, bytes)))), str_fmt
1172 elif fmt
[-1] == 'j': # json
1173 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
, indent
=4 if '#' in flags
else None), str_fmt
1174 elif fmt
[-1] == 'q': # quoted
1175 value
= map(str, variadic(value
) if '#' in flags
else [value
])
1176 value
, fmt
= ' '.join(map(compat_shlex_quote
, value
)), str_fmt
1177 elif fmt
[-1] == 'B': # bytes
1178 value
= f
'%{str_fmt}'.encode() % str(value
).encode('utf-8')
1179 value
, fmt
= value
.decode('utf-8', 'ignore'), 's'
1180 elif fmt
[-1] == 'U': # unicode normalized
1181 value
, fmt
= unicodedata
.normalize(
1182 # "+" = compatibility equivalence, "#" = NFD
1183 'NF%s%s' % ('K' if '+' in flags
else '', 'D' if '#' in flags
else 'C'),
1185 elif fmt
[-1] == 'D': # decimal suffix
1186 num_fmt
, fmt
= fmt
[:-1].replace('#', ''), 's'
1187 value
= format_decimal_suffix(value
, f
'%{num_fmt}f%s' if num_fmt
else '%d%s',
1188 factor
=1024 if '#' in flags
else 1000)
1189 elif fmt
[-1] == 'S': # filename sanitization
1190 value
, fmt
= filename_sanitizer(initial_field
, value
, restricted
='#' in flags
), str_fmt
1191 elif fmt
[-1] == 'c':
1193 value
= str(value
)[0]
1196 elif fmt
[-1] not in 'rs': # numeric
1197 value
= float_or_none(value
)
1199 value
, fmt
= default
, 's'
1203 # If value is an object, sanitize might convert it to a string
1204 # So we convert it to repr first
1205 value
, fmt
= repr(value
), str_fmt
1206 if fmt
[-1] in 'csr':
1207 value
= sanitizer(initial_field
, value
)
1209 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1210 TMPL_DICT
[key
] = value
1211 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1213 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1215 def evaluate_outtmpl(self
, outtmpl
, info_dict
, *args
, **kwargs
):
1216 outtmpl
, info_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, *args
, **kwargs
)
1217 return self
.escape_outtmpl(outtmpl
) % info_dict
1219 def _prepare_filename(self
, info_dict
, *, outtmpl
=None, tmpl_type
=None):
1220 assert None in (outtmpl
, tmpl_type
), 'outtmpl and tmpl_type are mutually exclusive'
1222 outtmpl
= self
.outtmpl_dict
.get(tmpl_type
or 'default', self
.outtmpl_dict
['default'])
1224 outtmpl
= self
._outtmpl
_expandpath
(outtmpl
)
1225 filename
= self
.evaluate_outtmpl(outtmpl
, info_dict
, True)
1229 if tmpl_type
in ('', 'temp'):
1230 final_ext
, ext
= self
.params
.get('final_ext'), info_dict
.get('ext')
1231 if final_ext
and ext
and final_ext
!= ext
and filename
.endswith(f
'.{final_ext}'):
1232 filename
= replace_extension(filename
, ext
, final_ext
)
1234 force_ext
= OUTTMPL_TYPES
[tmpl_type
]
1236 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1238 # https://github.com/blackjack4494/youtube-dlc/issues/85
1239 trim_file_name
= self
.params
.get('trim_file_name', False)
1241 no_ext
, *ext
= filename
.rsplit('.', 2)
1242 filename
= join_nonempty(no_ext
[:trim_file_name
], *ext
, delim
='.')
1245 except ValueError as err
:
1246 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1249 def prepare_filename(self
, info_dict
, dir_type
='', *, outtmpl
=None, warn
=False):
1250 """Generate the output filename"""
1252 assert not dir_type
, 'outtmpl and dir_type are mutually exclusive'
1254 filename
= self
._prepare
_filename
(info_dict
, tmpl_type
=dir_type
, outtmpl
=outtmpl
)
1255 if not filename
and dir_type
not in ('', 'temp'):
1259 if not self
.params
.get('paths'):
1261 elif filename
== '-':
1262 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1263 elif os
.path
.isabs(filename
):
1264 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1265 if filename
== '-' or not filename
:
1268 return self
.get_output_path(dir_type
, filename
)
1270 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1271 """ Returns None if the file should be downloaded """
1273 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1276 if 'title' in info_dict
:
1277 # This can happen when we're just evaluating the playlist
1278 title
= info_dict
['title']
1279 matchtitle
= self
.params
.get('matchtitle', False)
1281 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1282 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1283 rejecttitle
= self
.params
.get('rejecttitle', False)
1285 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1286 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1287 date
= info_dict
.get('upload_date')
1288 if date
is not None:
1289 dateRange
= self
.params
.get('daterange', DateRange())
1290 if date
not in dateRange
:
1291 return f
'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
1292 view_count
= info_dict
.get('view_count')
1293 if view_count
is not None:
1294 min_views
= self
.params
.get('min_views')
1295 if min_views
is not None and view_count
< min_views
:
1296 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1297 max_views
= self
.params
.get('max_views')
1298 if max_views
is not None and view_count
> max_views
:
1299 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1300 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1301 return 'Skipping "%s" because it is age restricted' % video_title
1303 match_filter
= self
.params
.get('match_filter')
1304 if match_filter
is not None:
1306 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1308 # For backward compatibility
1309 ret
= None if incomplete
else match_filter(info_dict
)
1314 if self
.in_download_archive(info_dict
):
1315 reason
= '%s has already been recorded in the archive' % video_title
1316 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1318 reason
= check_filter()
1319 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1320 if reason
is not None:
1322 self
.to_screen('[download] ' + reason
)
1323 if self
.params
.get(break_opt
, False):
1328 def add_extra_info(info_dict
, extra_info
):
1329 '''Set the keys from extra_info in info dict if they are missing'''
1330 for key
, value
in extra_info
.items():
1331 info_dict
.setdefault(key
, value
)
1333 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
=None,
1334 process
=True, force_generic_extractor
=False):
1336 Return a list with a dictionary for each video extracted.
1339 url -- URL to extract
1342 download -- whether to download videos during extraction
1343 ie_key -- extractor key hint
1344 extra_info -- dictionary containing the extra values to add to each result
1345 process -- whether to resolve all unresolved references (URLs, playlist items),
1346 must be True for download to work.
1347 force_generic_extractor -- force using the generic extractor
1350 if extra_info
is None:
1353 if not ie_key
and force_generic_extractor
:
1357 ies
= {ie_key: self._get_info_extractor_class(ie_key)}
1361 for ie_key
, ie
in ies
.items():
1362 if not ie
.suitable(url
):
1365 if not ie
.working():
1366 self
.report_warning('The program functionality for this site has been marked as broken, '
1367 'and will probably not work.')
1369 temp_id
= ie
.get_temp_id(url
)
1370 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1371 self
.to_screen(f
'[{ie_key}] {temp_id}: has already been recorded in the archive')
1372 if self
.params
.get('break_on_existing', False):
1373 raise ExistingVideoReached()
1375 return self
.__extract
_info
(url
, self
.get_info_extractor(ie_key
), download
, extra_info
, process
)
1377 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1379 def __handle_extraction_exceptions(func
):
1380 @functools.wraps(func
)
1381 def wrapper(self
, *args
, **kwargs
):
1384 return func(self
, *args
, **kwargs
)
1385 except (DownloadCancelled
, LazyList
.IndexError, PagedList
.IndexError):
1387 except ReExtractInfo
as e
:
1389 self
.to_screen(f
'{e}; Re-extracting data')
1391 self
.to_stderr('\r')
1392 self
.report_warning(f
'{e}; Re-extracting data')
1394 except GeoRestrictedError
as e
:
1397 msg
+= '\nThis video is available in %s.' % ', '.join(
1398 map(ISO3166Utils
.short2full
, e
.countries
))
1399 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1400 self
.report_error(msg
)
1401 except ExtractorError
as e
: # An error we somewhat expected
1402 self
.report_error(str(e
), e
.format_traceback())
1403 except Exception as e
:
1404 if self
.params
.get('ignoreerrors'):
1405 self
.report_error(str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1411 def _wait_for_video(self
, ie_result
):
1412 if (not self
.params
.get('wait_for_video')
1413 or ie_result
.get('_type', 'video') != 'video'
1414 or ie_result
.get('formats') or ie_result
.get('url')):
1417 format_dur
= lambda dur
: '%02d:%02d:%02d' % timetuple_from_msec(dur
* 1000)[:-1]
1422 self
.to_screen(msg
+ ' ' * (len(last_msg
) - len(msg
)) + '\r', skip_eol
=True)
1425 min_wait
, max_wait
= self
.params
.get('wait_for_video')
1426 diff
= try_get(ie_result
, lambda x
: x
['release_timestamp'] - time
.time())
1427 if diff
is None and ie_result
.get('live_status') == 'is_upcoming':
1428 diff
= round(random
.uniform(min_wait
, max_wait
) if (max_wait
and min_wait
) else (max_wait
or min_wait
), 0)
1429 self
.report_warning('Release time of video is not known')
1430 elif (diff
or 0) <= 0:
1431 self
.report_warning('Video should already be available according to extracted info')
1432 diff
= min(max(diff
or 0, min_wait
or 0), max_wait
or float('inf'))
1433 self
.to_screen(f
'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1435 wait_till
= time
.time() + diff
1438 diff
= wait_till
- time
.time()
1441 raise ReExtractInfo('[wait] Wait period ended', expected
=True)
1442 progress(f
'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1444 except KeyboardInterrupt:
1446 raise ReExtractInfo('[wait] Interrupted by user', expected
=True)
1447 except BaseException
as e
:
1448 if not isinstance(e
, ReExtractInfo
):
1452 @__handle_extraction_exceptions
1453 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1454 ie_result
= ie
.extract(url
)
1455 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1457 if isinstance(ie_result
, list):
1458 # Backwards compatibility: old IE result format
1460 '_type': 'compat_list',
1461 'entries': ie_result
,
1463 if extra_info
.get('original_url'):
1464 ie_result
.setdefault('original_url', extra_info
['original_url'])
1465 self
.add_default_extra_info(ie_result
, ie
, url
)
1467 self
._wait
_for
_video
(ie_result
)
1468 return self
.process_ie_result(ie_result
, download
, extra_info
)
1472 def add_default_extra_info(self
, ie_result
, ie
, url
):
1474 self
.add_extra_info(ie_result
, {
1476 'original_url': url
,
1478 webpage_url
= ie_result
.get('webpage_url')
1480 self
.add_extra_info(ie_result
, {
1481 'webpage_url_basename': url_basename(webpage_url
),
1482 'webpage_url_domain': get_domain(webpage_url
),
1485 self
.add_extra_info(ie_result
, {
1486 'extractor': ie
.IE_NAME
,
1487 'extractor_key': ie
.ie_key(),
1490 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1492 Take the result of the ie(may be modified) and resolve all unresolved
1493 references (URLs, playlist items).
1495 It will also download the videos if 'download'.
1496 Returns the resolved ie_result.
1498 if extra_info
is None:
1500 result_type
= ie_result
.get('_type', 'video')
1502 if result_type
in ('url', 'url_transparent'):
1503 ie_result
['url'] = sanitize_url(ie_result
['url'])
1504 if ie_result
.get('original_url'):
1505 extra_info
.setdefault('original_url', ie_result
['original_url'])
1507 extract_flat
= self
.params
.get('extract_flat', False)
1508 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1509 or extract_flat
is True):
1510 info_copy
= ie_result
.copy()
1511 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1512 if ie
and not ie_result
.get('id'):
1513 info_copy
['id'] = ie
.get_temp_id(ie_result
['url'])
1514 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1515 self
.add_extra_info(info_copy
, extra_info
)
1516 info_copy
, _
= self
.pre_process(info_copy
)
1517 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1518 if self
.params
.get('force_write_download_archive', False):
1519 self
.record_download_archive(info_copy
)
1522 if result_type
== 'video':
1523 self
.add_extra_info(ie_result
, extra_info
)
1524 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1525 additional_urls
= (ie_result
or {}).get('additional_urls')
1527 # TODO: Improve MetadataParserPP to allow setting a list
1528 if isinstance(additional_urls
, compat_str
):
1529 additional_urls
= [additional_urls
]
1531 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1532 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1533 ie_result
['additional_entries'] = [
1535 url
, download
, extra_info
=extra_info
,
1536 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1537 for url
in additional_urls
1540 elif result_type
== 'url':
1541 # We have to add extra_info to the results because it may be
1542 # contained in a playlist
1543 return self
.extract_info(
1544 ie_result
['url'], download
,
1545 ie_key
=ie_result
.get('ie_key'),
1546 extra_info
=extra_info
)
1547 elif result_type
== 'url_transparent':
1548 # Use the information from the embedding page
1549 info
= self
.extract_info(
1550 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1551 extra_info
=extra_info
, download
=False, process
=False)
1553 # extract_info may return None when ignoreerrors is enabled and
1554 # extraction failed with an error, don't crash and return early
1559 new_result
= info
.copy()
1560 new_result
.update(filter_dict(ie_result
, lambda k
, v
: (
1561 v
is not None and k
not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'}
)))
1563 # Extracted info may not be a video result (i.e.
1564 # info.get('_type', 'video') != video) but rather an url or
1565 # url_transparent. In such cases outer metadata (from ie_result)
1566 # should be propagated to inner one (info). For this to happen
1567 # _type of info should be overridden with url_transparent. This
1568 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1569 if new_result
.get('_type') == 'url':
1570 new_result
['_type'] = 'url_transparent'
1572 return self
.process_ie_result(
1573 new_result
, download
=download
, extra_info
=extra_info
)
1574 elif result_type
in ('playlist', 'multi_video'):
1575 # Protect from infinite recursion due to recursively nested playlists
1576 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1577 webpage_url
= ie_result
['webpage_url']
1578 if webpage_url
in self
._playlist
_urls
:
1580 '[download] Skipping already downloaded playlist: %s'
1581 % ie_result
.get('title') or ie_result
.get('id'))
1584 self
._playlist
_level
+= 1
1585 self
._playlist
_urls
.add(webpage_url
)
1586 self
._fill
_common
_fields
(ie_result
, False)
1587 self
._sanitize
_thumbnails
(ie_result
)
1589 return self
.__process
_playlist
(ie_result
, download
)
1591 self
._playlist
_level
-= 1
1592 if not self
._playlist
_level
:
1593 self
._playlist
_urls
.clear()
1594 elif result_type
== 'compat_list':
1595 self
.report_warning(
1596 'Extractor %s returned a compat_list result. '
1597 'It needs to be updated.' % ie_result
.get('extractor'))
1600 self
.add_extra_info(r
, {
1601 'extractor': ie_result
['extractor'],
1602 'webpage_url': ie_result
['webpage_url'],
1603 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1604 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1605 'extractor_key': ie_result
['extractor_key'],
1608 ie_result
['entries'] = [
1609 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1610 for r
in ie_result
['entries']
1614 raise Exception('Invalid result type: %s' % result_type
)
1616 def _ensure_dir_exists(self
, path
):
1617 return make_dir(path
, self
.report_error
)
1620 def _playlist_infodict(ie_result
, **kwargs
):
1623 'playlist': ie_result
.get('title') or ie_result
.get('id'),
1624 'playlist_id': ie_result
.get('id'),
1625 'playlist_title': ie_result
.get('title'),
1626 'playlist_uploader': ie_result
.get('uploader'),
1627 'playlist_uploader_id': ie_result
.get('uploader_id'),
1628 'playlist_index': 0,
1632 def __process_playlist(self
, ie_result
, download
):
1633 # We process each entry in the playlist
1634 playlist
= ie_result
.get('title') or ie_result
.get('id')
1635 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1637 if 'entries' not in ie_result
:
1638 raise EntryNotInPlaylist('There are no entries')
1640 MissingEntry
= object()
1641 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1642 if incomplete_entries
:
1643 def fill_missing_entries(entries
, indices
):
1644 ret
= [MissingEntry
] * max(indices
)
1645 for i
, entry
in zip(indices
, entries
):
1648 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1650 playlist_results
= []
1652 playliststart
= self
.params
.get('playliststart', 1)
1653 playlistend
= self
.params
.get('playlistend')
1654 # For backwards compatibility, interpret -1 as whole list
1655 if playlistend
== -1:
1658 playlistitems_str
= self
.params
.get('playlist_items')
1659 playlistitems
= None
1660 if playlistitems_str
is not None:
1661 def iter_playlistitems(format
):
1662 for string_segment
in format
.split(','):
1663 if '-' in string_segment
:
1664 start
, end
= string_segment
.split('-')
1665 for item
in range(int(start
), int(end
) + 1):
1668 yield int(string_segment
)
1669 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1671 ie_entries
= ie_result
['entries']
1672 if isinstance(ie_entries
, list):
1673 playlist_count
= len(ie_entries
)
1674 msg
= f
'Collected {playlist_count} videos; downloading %d of them'
1675 ie_result
['playlist_count'] = ie_result
.get('playlist_count') or playlist_count
1678 return ie_entries
[i
- 1]
1680 msg
= 'Downloading %d videos'
1681 if not isinstance(ie_entries
, (PagedList
, LazyList
)):
1682 ie_entries
= LazyList(ie_entries
)
1683 elif isinstance(ie_entries
, InAdvancePagedList
):
1684 if ie_entries
._pagesize
== 1:
1685 playlist_count
= ie_entries
._pagecount
1688 return YoutubeDL
.__handle
_extraction
_exceptions
(
1689 lambda self
, i
: ie_entries
[i
- 1]
1692 entries
, broken
= [], False
1693 items
= playlistitems
if playlistitems
is not None else itertools
.count(playliststart
)
1697 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1701 entry
= get_entry(i
)
1702 if entry
is MissingEntry
:
1703 raise EntryNotInPlaylist()
1704 except (IndexError, EntryNotInPlaylist
):
1705 if incomplete_entries
:
1706 raise EntryNotInPlaylist(f
'Entry {i} cannot be found')
1707 elif not playlistitems
:
1709 entries
.append(entry
)
1711 if entry
is not None:
1712 # TODO: Add auto-generated fields
1713 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1714 except (ExistingVideoReached
, RejectedVideoReached
):
1717 ie_result
['entries'] = entries
1719 # Save playlist_index before re-ordering
1721 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1722 for i
, entry
in enumerate(entries
, 1)
1723 if entry
is not None]
1724 n_entries
= len(entries
)
1726 if not (ie_result
.get('playlist_count') or broken
or playlistitems
or playlistend
):
1727 ie_result
['playlist_count'] = n_entries
1729 if not playlistitems
and (playliststart
!= 1 or playlistend
):
1730 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1731 ie_result
['requested_entries'] = playlistitems
1733 _infojson_written
= False
1734 write_playlist_files
= self
.params
.get('allow_playlist_files', True)
1735 if write_playlist_files
and self
.params
.get('list_thumbnails'):
1736 self
.list_thumbnails(ie_result
)
1737 if write_playlist_files
and not self
.params
.get('simulate'):
1738 ie_copy
= self
._playlist
_infodict
(ie_result
, n_entries
=n_entries
)
1739 _infojson_written
= self
._write
_info
_json
(
1740 'playlist', ie_result
, self
.prepare_filename(ie_copy
, 'pl_infojson'))
1741 if _infojson_written
is None:
1743 if self
._write
_description
('playlist', ie_result
,
1744 self
.prepare_filename(ie_copy
, 'pl_description')) is None:
1746 # TODO: This should be passed to ThumbnailsConvertor if necessary
1747 self
._write
_thumbnails
('playlist', ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1749 if self
.params
.get('playlistreverse', False):
1750 entries
= entries
[::-1]
1751 if self
.params
.get('playlistrandom', False):
1752 random
.shuffle(entries
)
1754 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1756 self
.to_screen(f
'[{ie_result["extractor"]}] playlist {playlist}: {msg % n_entries}')
1758 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1759 for i
, entry_tuple
in enumerate(entries
, 1):
1760 playlist_index
, entry
= entry_tuple
1761 if 'playlist-index' in self
.params
.get('compat_opts', []):
1762 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1
1763 self
.to_screen(f
'[download] Downloading video {i} of {n_entries}')
1764 # This __x_forwarded_for_ip thing is a bit ugly but requires
1767 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1769 'n_entries': n_entries
,
1770 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1771 'playlist_count': ie_result
.get('playlist_count'),
1772 'playlist_index': playlist_index
,
1773 'playlist_autonumber': i
,
1774 'playlist': playlist
,
1775 'playlist_id': ie_result
.get('id'),
1776 'playlist_title': ie_result
.get('title'),
1777 'playlist_uploader': ie_result
.get('uploader'),
1778 'playlist_uploader_id': ie_result
.get('uploader_id'),
1779 'extractor': ie_result
['extractor'],
1780 'webpage_url': ie_result
['webpage_url'],
1781 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1782 'webpage_url_domain': get_domain(ie_result
['webpage_url']),
1783 'extractor_key': ie_result
['extractor_key'],
1786 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1789 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1790 if not entry_result
:
1792 if failures
>= max_failures
:
1794 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1796 playlist_results
.append(entry_result
)
1797 ie_result
['entries'] = playlist_results
1799 # Write the updated info to json
1800 if _infojson_written
is True and self
._write
_info
_json
(
1801 'updated playlist', ie_result
,
1802 self
.prepare_filename(ie_copy
, 'pl_infojson'), overwrite
=True) is None:
1805 ie_result
= self
.run_all_pps('playlist', ie_result
)
1806 self
.to_screen(f
'[download] Finished downloading playlist: {playlist}')
1809 @__handle_extraction_exceptions
1810 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1811 return self
.process_ie_result(
1812 entry
, download
=download
, extra_info
=extra_info
)
1814 def _build_format_filter(self
, filter_spec
):
1815 " Returns a function to filter the formats according to the filter_spec "
1825 operator_rex
= re
.compile(r
'''(?x)\s*
1826 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1827 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1828 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1829 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1830 m
= operator_rex
.fullmatch(filter_spec
)
1833 comparison_value
= int(m
.group('value'))
1835 comparison_value
= parse_filesize(m
.group('value'))
1836 if comparison_value
is None:
1837 comparison_value
= parse_filesize(m
.group('value') + 'B')
1838 if comparison_value
is None:
1840 'Invalid value %r in format specification %r' % (
1841 m
.group('value'), filter_spec
))
1842 op
= OPERATORS
[m
.group('op')]
1847 '^=': lambda attr
, value
: attr
.startswith(value
),
1848 '$=': lambda attr
, value
: attr
.endswith(value
),
1849 '*=': lambda attr
, value
: value
in attr
,
1850 '~=': lambda attr
, value
: value
.search(attr
) is not None
1852 str_operator_rex
= re
.compile(r
'''(?x)\s*
1853 (?P<key>[a-zA-Z0-9._-]+)\s*
1854 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1856 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1857 (?(quote)(?P=quote))\s*
1858 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1859 m
= str_operator_rex
.fullmatch(filter_spec
)
1861 if m
.group('op') == '~=':
1862 comparison_value
= re
.compile(m
.group('value'))
1864 comparison_value
= re
.sub(r
'''\\([\\"'])''', r
'\1', m
.group('value'))
1865 str_op
= STR_OPERATORS
[m
.group('op')]
1866 if m
.group('negation'):
1867 op
= lambda attr
, value
: not str_op(attr
, value
)
1872 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1875 actual_value
= f
.get(m
.group('key'))
1876 if actual_value
is None:
1877 return m
.group('none_inclusive')
1878 return op(actual_value
, comparison_value
)
1881 def _check_formats(self
, formats
):
1883 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1884 path
= self
.get_output_path('temp')
1885 if not self
._ensure
_dir
_exists
(f
'{path}/'):
1887 temp_file
= tempfile
.NamedTemporaryFile(suffix
='.tmp', delete
=False, dir=path
or None)
1890 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1891 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1894 if os
.path
.exists(temp_file
.name
):
1896 os
.remove(temp_file
.name
)
1898 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1902 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1904 def _default_format_spec(self
, info_dict
, download
=True):
1907 merger
= FFmpegMergerPP(self
)
1908 return merger
.available
and merger
.can_merge()
1911 not self
.params
.get('simulate')
1915 or info_dict
.get('is_live', False)
1916 or self
.outtmpl_dict
['default'] == '-'))
1919 or self
.params
.get('allow_multiple_audio_streams', False)
1920 or 'format-spec' in self
.params
.get('compat_opts', []))
1923 'best/bestvideo+bestaudio' if prefer_best
1924 else 'bestvideo*+bestaudio/best' if not compat
1925 else 'bestvideo+bestaudio/best')
1927 def build_format_selector(self
, format_spec
):
1928 def syntax_error(note
, start
):
1930 'Invalid format specification: '
1931 '{}\n\t{}\n\t{}^'.format(note
, format_spec
, ' ' * start
[1]))
1932 return SyntaxError(message
)
1934 PICKFIRST
= 'PICKFIRST'
1938 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1940 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1941 'video': self
.params
.get('allow_multiple_video_streams', False)}
1943 check_formats
= self
.params
.get('check_formats') == 'selected'
1945 def _parse_filter(tokens
):
1947 for type, string
, start
, _
, _
in tokens
:
1948 if type == tokenize
.OP
and string
== ']':
1949 return ''.join(filter_parts
)
1951 filter_parts
.append(string
)
1953 def _remove_unused_ops(tokens
):
1954 # Remove operators that we don't use and join them with the surrounding strings
1955 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1956 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1957 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1958 for type, string
, start
, end
, line
in tokens
:
1959 if type == tokenize
.OP
and string
== '[':
1961 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1963 yield type, string
, start
, end
, line
1964 # everything inside brackets will be handled by _parse_filter
1965 for type, string
, start
, end
, line
in tokens
:
1966 yield type, string
, start
, end
, line
1967 if type == tokenize
.OP
and string
== ']':
1969 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1971 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1973 yield type, string
, start
, end
, line
1974 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1976 last_string
= string
1980 last_string
+= string
1982 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1984 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1986 current_selector
= None
1987 for type, string
, start
, _
, _
in tokens
:
1988 # ENCODING is only defined in python 3.x
1989 if type == getattr(tokenize
, 'ENCODING', None):
1991 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1992 current_selector
= FormatSelector(SINGLE
, string
, [])
1993 elif type == tokenize
.OP
:
1995 if not inside_group
:
1996 # ')' will be handled by the parentheses group
1997 tokens
.restore_last_token()
1999 elif inside_merge
and string
in ['/', ',']:
2000 tokens
.restore_last_token()
2002 elif inside_choice
and string
== ',':
2003 tokens
.restore_last_token()
2006 if not current_selector
:
2007 raise syntax_error('"," must follow a format selector', start
)
2008 selectors
.append(current_selector
)
2009 current_selector
= None
2011 if not current_selector
:
2012 raise syntax_error('"/" must follow a format selector', start
)
2013 first_choice
= current_selector
2014 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
2015 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
2017 if not current_selector
:
2018 current_selector
= FormatSelector(SINGLE
, 'best', [])
2019 format_filter
= _parse_filter(tokens
)
2020 current_selector
.filters
.append(format_filter
)
2022 if current_selector
:
2023 raise syntax_error('Unexpected "("', start
)
2024 group
= _parse_format_selection(tokens
, inside_group
=True)
2025 current_selector
= FormatSelector(GROUP
, group
, [])
2027 if not current_selector
:
2028 raise syntax_error('Unexpected "+"', start
)
2029 selector_1
= current_selector
2030 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
2032 raise syntax_error('Expected a selector', start
)
2033 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
2035 raise syntax_error(f
'Operator not recognized: "{string}"', start
)
2036 elif type == tokenize
.ENDMARKER
:
2038 if current_selector
:
2039 selectors
.append(current_selector
)
2042 def _merge(formats_pair
):
2043 format_1
, format_2
= formats_pair
2046 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
2047 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
2049 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
2050 get_no_more
= {'video': False, 'audio': False}
2051 for (i
, fmt_info
) in enumerate(formats_info
):
2052 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
2055 for aud_vid
in ['audio', 'video']:
2056 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
2057 if get_no_more
[aud_vid
]:
2060 get_no_more
[aud_vid
] = True
2062 if len(formats_info
) == 1:
2063 return formats_info
[0]
2065 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
2066 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
2068 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
2069 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
2071 output_ext
= self
.params
.get('merge_output_format')
2074 output_ext
= the_only_video
['ext']
2075 elif the_only_audio
and not video_fmts
:
2076 output_ext
= the_only_audio
['ext']
2080 filtered
= lambda *keys
: filter(None, (traverse_obj(fmt
, *keys
) for fmt
in formats_info
))
2083 'requested_formats': formats_info
,
2084 'format': '+'.join(filtered('format')),
2085 'format_id': '+'.join(filtered('format_id')),
2087 'protocol': '+'.join(map(determine_protocol
, formats_info
)),
2088 'language': '+'.join(orderedSet(filtered('language'))) or None,
2089 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2090 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2091 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2096 'width': the_only_video
.get('width'),
2097 'height': the_only_video
.get('height'),
2098 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
2099 'fps': the_only_video
.get('fps'),
2100 'dynamic_range': the_only_video
.get('dynamic_range'),
2101 'vcodec': the_only_video
.get('vcodec'),
2102 'vbr': the_only_video
.get('vbr'),
2103 'stretched_ratio': the_only_video
.get('stretched_ratio'),
2108 'acodec': the_only_audio
.get('acodec'),
2109 'abr': the_only_audio
.get('abr'),
2110 'asr': the_only_audio
.get('asr'),
2115 def _check_formats(formats
):
2116 if not check_formats
:
2119 yield from self
._check
_formats
(formats
)
2121 def _build_selector_function(selector
):
2122 if isinstance(selector
, list): # ,
2123 fs
= [_build_selector_function(s
) for s
in selector
]
2125 def selector_function(ctx
):
2128 return selector_function
2130 elif selector
.type == GROUP
: # ()
2131 selector_function
= _build_selector_function(selector
.selector
)
2133 elif selector
.type == PICKFIRST
: # /
2134 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
2136 def selector_function(ctx
):
2138 picked_formats
= list(f(ctx
))
2140 return picked_formats
2143 elif selector
.type == MERGE
: # +
2144 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
2146 def selector_function(ctx
):
2147 for pair
in itertools
.product(selector_1(ctx
), selector_2(ctx
)):
2150 elif selector
.type == SINGLE
: # atom
2151 format_spec
= selector
.selector
or 'best'
2153 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2154 if format_spec
== 'all':
2155 def selector_function(ctx
):
2156 yield from _check_formats(ctx
['formats'][::-1])
2157 elif format_spec
== 'mergeall':
2158 def selector_function(ctx
):
2159 formats
= list(_check_formats(
2160 f
for f
in ctx
['formats'] if f
.get('vcodec') != 'none' or f
.get('acodec') != 'none'))
2163 merged_format
= formats
[-1]
2164 for f
in formats
[-2::-1]:
2165 merged_format
= _merge((merged_format
, f
))
2169 format_fallback
, seperate_fallback
, format_reverse
, format_idx
= False, None, True, 1
2171 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2173 if mobj
is not None:
2174 format_idx
= int_or_none(mobj
.group('n'), default
=1)
2175 format_reverse
= mobj
.group('bw')[0] == 'b'
2176 format_type
= (mobj
.group('type') or [None])[0]
2177 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
2178 format_modified
= mobj
.group('mod') is not None
2180 format_fallback
= not format_type
and not format_modified
# for b, w
2182 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
2183 if format_type
and format_modified
# bv*, ba*, wv*, wa*
2184 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
2185 if format_type
# bv, ba, wv, wa
2186 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
2187 if not format_modified
# b, w
2188 else lambda f
: True) # b*, w*
2189 filter_f
= lambda f
: _filter_f(f
) and (
2190 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
2192 if format_spec
in self
._format
_selection
_exts
['audio']:
2193 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none'
2194 elif format_spec
in self
._format
_selection
_exts
['video']:
2195 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') != 'none' and f
.get('vcodec') != 'none'
2196 seperate_fallback
= lambda f
: f
.get('ext') == format_spec
and f
.get('vcodec') != 'none'
2197 elif format_spec
in self
._format
_selection
_exts
['storyboards']:
2198 filter_f
= lambda f
: f
.get('ext') == format_spec
and f
.get('acodec') == 'none' and f
.get('vcodec') == 'none'
2200 filter_f
= lambda f
: f
.get('format_id') == format_spec
# id
2202 def selector_function(ctx
):
2203 formats
= list(ctx
['formats'])
2204 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
2206 if format_fallback
and ctx
['incomplete_formats']:
2207 # for extractors with incomplete formats (audio only (soundcloud)
2208 # or video only (imgur)) best/worst will fallback to
2209 # best/worst {video,audio}-only format
2211 elif seperate_fallback
and not ctx
['has_merged_format']:
2212 # for compatibility with youtube-dl when there is no pre-merged format
2213 matches
= list(filter(seperate_fallback
, formats
))
2214 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
2216 yield matches
[format_idx
- 1]
2217 except LazyList
.IndexError:
2220 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
2222 def final_selector(ctx
):
2223 ctx_copy
= dict(ctx
)
2224 for _filter
in filters
:
2225 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
2226 return selector_function(ctx_copy
)
2227 return final_selector
2229 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
2231 tokens
= list(_remove_unused_ops(tokenize
.tokenize(stream
.readline
)))
2232 except tokenize
.TokenError
:
2233 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
2235 class TokenIterator
:
2236 def __init__(self
, tokens
):
2237 self
.tokens
= tokens
2244 if self
.counter
>= len(self
.tokens
):
2245 raise StopIteration()
2246 value
= self
.tokens
[self
.counter
]
2252 def restore_last_token(self
):
2255 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
2256 return _build_selector_function(parsed_selector
)
2258 def _calc_headers(self
, info_dict
):
2259 res
= merge_headers(self
.params
['http_headers'], info_dict
.get('http_headers') or {})
2261 cookies
= self
._calc
_cookies
(info_dict
)
2263 res
['Cookie'] = cookies
2265 if 'X-Forwarded-For' not in res
:
2266 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2267 if x_forwarded_for_ip
:
2268 res
['X-Forwarded-For'] = x_forwarded_for_ip
2272 def _calc_cookies(self
, info_dict
):
2273 pr
= sanitized_Request(info_dict
['url'])
2274 self
.cookiejar
.add_cookie_header(pr
)
2275 return pr
.get_header('Cookie')
2277 def _sort_thumbnails(self
, thumbnails
):
2278 thumbnails
.sort(key
=lambda t
: (
2279 t
.get('preference') if t
.get('preference') is not None else -1,
2280 t
.get('width') if t
.get('width') is not None else -1,
2281 t
.get('height') if t
.get('height') is not None else -1,
2282 t
.get('id') if t
.get('id') is not None else '',
2285 def _sanitize_thumbnails(self
, info_dict
):
2286 thumbnails
= info_dict
.get('thumbnails')
2287 if thumbnails
is None:
2288 thumbnail
= info_dict
.get('thumbnail')
2290 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2294 def check_thumbnails(thumbnails
):
2295 for t
in thumbnails
:
2296 self
.to_screen(f
'[info] Testing thumbnail {t["id"]}')
2298 self
.urlopen(HEADRequest(t
['url']))
2299 except network_exceptions
as err
:
2300 self
.to_screen(f
'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2304 self
._sort
_thumbnails
(thumbnails
)
2305 for i
, t
in enumerate(thumbnails
):
2306 if t
.get('id') is None:
2308 if t
.get('width') and t
.get('height'):
2309 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2310 t
['url'] = sanitize_url(t
['url'])
2312 if self
.params
.get('check_formats') is True:
2313 info_dict
['thumbnails'] = LazyList(check_thumbnails(thumbnails
[::-1]), reverse
=True)
2315 info_dict
['thumbnails'] = thumbnails
2317 def _fill_common_fields(self
, info_dict
, is_video
=True):
2318 # TODO: move sanitization here
2320 # playlists are allowed to lack "title"
2321 info_dict
['fulltitle'] = info_dict
.get('title')
2322 if 'title' not in info_dict
:
2323 raise ExtractorError('Missing "title" field in extractor result',
2324 video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2325 elif not info_dict
.get('title'):
2326 self
.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2327 info_dict
['title'] = f
'{info_dict["extractor"]} video #{info_dict["id"]}'
2329 if info_dict
.get('duration') is not None:
2330 info_dict
['duration_string'] = formatSeconds(info_dict
['duration'])
2332 for ts_key
, date_key
in (
2333 ('timestamp', 'upload_date'),
2334 ('release_timestamp', 'release_date'),
2335 ('modified_timestamp', 'modified_date'),
2337 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2338 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2339 # see http://bugs.python.org/issue1646728)
2341 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2342 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2343 except (ValueError, OverflowError, OSError):
2346 live_keys
= ('is_live', 'was_live')
2347 live_status
= info_dict
.get('live_status')
2348 if live_status
is None:
2349 for key
in live_keys
:
2350 if info_dict
.get(key
) is False:
2352 if info_dict
.get(key
):
2355 if all(info_dict
.get(key
) is False for key
in live_keys
):
2356 live_status
= 'not_live'
2358 info_dict
['live_status'] = live_status
2359 for key
in live_keys
:
2360 if info_dict
.get(key
) is None:
2361 info_dict
[key
] = (live_status
== key
)
2363 # Auto generate title fields corresponding to the *_number fields when missing
2364 # in order to always have clean titles. This is very common for TV series.
2365 for field
in ('chapter', 'season', 'episode'):
2366 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2367 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2369 def process_video_result(self
, info_dict
, download
=True):
2370 assert info_dict
.get('_type', 'video') == 'video'
2371 self
._num
_videos
+= 1
2373 if 'id' not in info_dict
:
2374 raise ExtractorError('Missing "id" field in extractor result', ie
=info_dict
['extractor'])
2375 elif not info_dict
.get('id'):
2376 raise ExtractorError('Extractor failed to obtain "id"', ie
=info_dict
['extractor'])
2378 def report_force_conversion(field
, field_not
, conversion
):
2379 self
.report_warning(
2380 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2381 % (field
, field_not
, conversion
))
2383 def sanitize_string_field(info
, string_field
):
2384 field
= info
.get(string_field
)
2385 if field
is None or isinstance(field
, compat_str
):
2387 report_force_conversion(string_field
, 'a string', 'string')
2388 info
[string_field
] = compat_str(field
)
2390 def sanitize_numeric_fields(info
):
2391 for numeric_field
in self
._NUMERIC
_FIELDS
:
2392 field
= info
.get(numeric_field
)
2393 if field
is None or isinstance(field
, (int, float)):
2395 report_force_conversion(numeric_field
, 'numeric', 'int')
2396 info
[numeric_field
] = int_or_none(field
)
2398 sanitize_string_field(info_dict
, 'id')
2399 sanitize_numeric_fields(info_dict
)
2400 if (info_dict
.get('duration') or 0) <= 0 and info_dict
.pop('duration', None):
2401 self
.report_warning('"duration" field is negative, there is an error in extractor')
2403 if 'playlist' not in info_dict
:
2404 # It isn't part of a playlist
2405 info_dict
['playlist'] = None
2406 info_dict
['playlist_index'] = None
2408 self
._sanitize
_thumbnails
(info_dict
)
2410 thumbnail
= info_dict
.get('thumbnail')
2411 thumbnails
= info_dict
.get('thumbnails')
2413 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2415 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2417 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2418 info_dict
['display_id'] = info_dict
['id']
2420 self
._fill
_common
_fields
(info_dict
)
2422 for cc_kind
in ('subtitles', 'automatic_captions'):
2423 cc
= info_dict
.get(cc_kind
)
2425 for _
, subtitle
in cc
.items():
2426 for subtitle_format
in subtitle
:
2427 if subtitle_format
.get('url'):
2428 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2429 if subtitle_format
.get('ext') is None:
2430 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2432 automatic_captions
= info_dict
.get('automatic_captions')
2433 subtitles
= info_dict
.get('subtitles')
2435 info_dict
['requested_subtitles'] = self
.process_subtitles(
2436 info_dict
['id'], subtitles
, automatic_captions
)
2438 if info_dict
.get('formats') is None:
2439 # There's only one format available
2440 formats
= [info_dict
]
2442 formats
= info_dict
['formats']
2444 info_dict
['__has_drm'] = any(f
.get('has_drm') for f
in formats
)
2445 if not self
.params
.get('allow_unplayable_formats'):
2446 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2447 if info_dict
['__has_drm'] and all(
2448 f
.get('acodec') == f
.get('vcodec') == 'none' for f
in formats
):
2449 self
.report_warning(
2450 'This video is DRM protected and only images are available for download. '
2451 'Use --list-formats to see them')
2453 get_from_start
= not info_dict
.get('is_live') or bool(self
.params
.get('live_from_start'))
2454 if not get_from_start
:
2455 info_dict
['title'] += ' ' + datetime
.datetime
.now().strftime('%Y-%m-%d %H:%M')
2456 if info_dict
.get('is_live') and formats
:
2457 formats
= [f
for f
in formats
if bool(f
.get('is_from_start')) == get_from_start
]
2458 if get_from_start
and not formats
:
2459 self
.raise_no_formats(info_dict
, msg
=(
2460 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2461 'If you want to download from the current time, use --no-live-from-start'))
2464 self
.raise_no_formats(info_dict
)
2466 def is_wellformed(f
):
2469 self
.report_warning(
2470 '"url" field is missing or empty - skipping format, '
2471 'there is an error in extractor')
2473 if isinstance(url
, bytes):
2474 sanitize_string_field(f
, 'url')
2477 # Filter out malformed formats for better extraction robustness
2478 formats
= list(filter(is_wellformed
, formats
))
2482 # We check that all the formats have the format and format_id fields
2483 for i
, format
in enumerate(formats
):
2484 sanitize_string_field(format
, 'format_id')
2485 sanitize_numeric_fields(format
)
2486 format
['url'] = sanitize_url(format
['url'])
2487 if not format
.get('format_id'):
2488 format
['format_id'] = compat_str(i
)
2490 # Sanitize format_id from characters used in format selector expression
2491 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2492 format_id
= format
['format_id']
2493 if format_id
not in formats_dict
:
2494 formats_dict
[format_id
] = []
2495 formats_dict
[format_id
].append(format
)
2497 # Make sure all formats have unique format_id
2498 common_exts
= set(itertools
.chain(*self
._format
_selection
_exts
.values()))
2499 for format_id
, ambiguous_formats
in formats_dict
.items():
2500 ambigious_id
= len(ambiguous_formats
) > 1
2501 for i
, format
in enumerate(ambiguous_formats
):
2503 format
['format_id'] = '%s-%d' % (format_id
, i
)
2504 if format
.get('ext') is None:
2505 format
['ext'] = determine_ext(format
['url']).lower()
2506 # Ensure there is no conflict between id and ext in format selection
2507 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2508 if format
['format_id'] != format
['ext'] and format
['format_id'] in common_exts
:
2509 format
['format_id'] = 'f%s' % format
['format_id']
2511 for i
, format
in enumerate(formats
):
2512 if format
.get('format') is None:
2513 format
['format'] = '{id} - {res}{note}'.format(
2514 id=format
['format_id'],
2515 res
=self
.format_resolution(format
),
2516 note
=format_field(format
, 'format_note', ' (%s)'),
2518 if format
.get('protocol') is None:
2519 format
['protocol'] = determine_protocol(format
)
2520 if format
.get('resolution') is None:
2521 format
['resolution'] = self
.format_resolution(format
, default
=None)
2522 if format
.get('dynamic_range') is None and format
.get('vcodec') != 'none':
2523 format
['dynamic_range'] = 'SDR'
2524 if (info_dict
.get('duration') and format
.get('tbr')
2525 and not format
.get('filesize') and not format
.get('filesize_approx')):
2526 format
['filesize_approx'] = info_dict
['duration'] * format
['tbr'] * (1024 / 8)
2528 # Add HTTP headers, so that external programs can use them from the
2530 full_format_info
= info_dict
.copy()
2531 full_format_info
.update(format
)
2532 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2533 # Remove private housekeeping stuff
2534 if '__x_forwarded_for_ip' in info_dict
:
2535 del info_dict
['__x_forwarded_for_ip']
2537 if self
.params
.get('check_formats') is True:
2538 formats
= LazyList(self
._check
_formats
(formats
[::-1]), reverse
=True)
2540 if not formats
or formats
[0] is not info_dict
:
2541 # only set the 'formats' fields if the original info_dict list them
2542 # otherwise we end up with a circular reference, the first (and unique)
2543 # element in the 'formats' field in info_dict is info_dict itself,
2544 # which can't be exported to json
2545 info_dict
['formats'] = formats
2547 info_dict
, _
= self
.pre_process(info_dict
)
2549 if self
._match
_entry
(info_dict
, incomplete
=self
._format
_fields
) is not None:
2552 self
.post_extract(info_dict
)
2553 info_dict
, _
= self
.pre_process(info_dict
, 'after_filter')
2555 # The pre-processors may have modified the formats
2556 formats
= info_dict
.get('formats', [info_dict
])
2558 list_only
= self
.params
.get('simulate') is None and (
2559 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2560 interactive_format_selection
= not list_only
and self
.format_selector
== '-'
2561 if self
.params
.get('list_thumbnails'):
2562 self
.list_thumbnails(info_dict
)
2563 if self
.params
.get('listsubtitles'):
2564 if 'automatic_captions' in info_dict
:
2565 self
.list_subtitles(
2566 info_dict
['id'], automatic_captions
, 'automatic captions')
2567 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2568 if self
.params
.get('listformats') or interactive_format_selection
:
2569 self
.list_formats(info_dict
)
2571 # Without this printing, -F --print-json will not work
2572 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2575 format_selector
= self
.format_selector
2576 if format_selector
is None:
2577 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2578 self
.write_debug('Default format spec: %s' % req_format
)
2579 format_selector
= self
.build_format_selector(req_format
)
2582 if interactive_format_selection
:
2584 self
._format
_screen
('\nEnter format selector: ', self
.Styles
.EMPHASIS
))
2586 format_selector
= self
.build_format_selector(req_format
)
2587 except SyntaxError as err
:
2588 self
.report_error(err
, tb
=False, is_error
=False)
2591 formats_to_download
= list(format_selector({
2593 'has_merged_format': any('none' not in (f
.get('acodec'), f
.get('vcodec')) for f
in formats
),
2594 'incomplete_formats': (
2595 # All formats are video-only or
2596 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2597 # all formats are audio-only
2598 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
)),
2600 if interactive_format_selection
and not formats_to_download
:
2601 self
.report_error('Requested format is not available', tb
=False, is_error
=False)
2605 if not formats_to_download
:
2606 if not self
.params
.get('ignore_no_formats_error'):
2607 raise ExtractorError(
2608 'Requested format is not available. Use --list-formats for a list of available formats',
2609 expected
=True, video_id
=info_dict
['id'], ie
=info_dict
['extractor'])
2610 self
.report_warning('Requested format is not available')
2611 # Process what we can, even without any available formats.
2612 formats_to_download
= [{}]
2614 best_format
= formats_to_download
[-1]
2618 f
'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2619 + ', '.join([f
['format_id'] for f
in formats_to_download
]))
2620 max_downloads_reached
= False
2621 for i
, fmt
in enumerate(formats_to_download
):
2622 formats_to_download
[i
] = new_info
= self
._copy
_infodict
(info_dict
)
2623 new_info
.update(fmt
)
2625 self
.process_info(new_info
)
2626 except MaxDownloadsReached
:
2627 max_downloads_reached
= True
2628 # Remove copied info
2629 for key
, val
in tuple(new_info
.items()):
2630 if info_dict
.get(key
) == val
:
2632 if max_downloads_reached
:
2635 write_archive
= {f.get('__write_download_archive', False) for f in formats_to_download}
2636 assert write_archive
.issubset({True, False, 'ignore'}
)
2637 if True in write_archive
and False not in write_archive
:
2638 self
.record_download_archive(info_dict
)
2640 info_dict
['requested_downloads'] = formats_to_download
2641 info_dict
= self
.run_all_pps('after_video', info_dict
)
2642 if max_downloads_reached
:
2643 raise MaxDownloadsReached()
2645 # We update the info dict with the selected best quality format (backwards compatibility)
2646 info_dict
.update(best_format
)
2649 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2650 """Select the requested subtitles and their format"""
2651 available_subs
, normal_sub_langs
= {}, []
2652 if normal_subtitles
and self
.params
.get('writesubtitles'):
2653 available_subs
.update(normal_subtitles
)
2654 normal_sub_langs
= tuple(normal_subtitles
.keys())
2655 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2656 for lang
, cap_info
in automatic_captions
.items():
2657 if lang
not in available_subs
:
2658 available_subs
[lang
] = cap_info
2660 if (not self
.params
.get('writesubtitles') and not
2661 self
.params
.get('writeautomaticsub') or not
2665 all_sub_langs
= tuple(available_subs
.keys())
2666 if self
.params
.get('allsubtitles', False):
2667 requested_langs
= all_sub_langs
2668 elif self
.params
.get('subtitleslangs', False):
2669 # A list is used so that the order of languages will be the same as
2670 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2671 requested_langs
= []
2672 for lang_re
in self
.params
.get('subtitleslangs'):
2673 discard
= lang_re
[0] == '-'
2675 lang_re
= lang_re
[1:]
2676 if lang_re
== 'all':
2678 requested_langs
= []
2680 requested_langs
.extend(all_sub_langs
)
2682 current_langs
= filter(re
.compile(lang_re
+ '$').match
, all_sub_langs
)
2684 for lang
in current_langs
:
2685 while lang
in requested_langs
:
2686 requested_langs
.remove(lang
)
2688 requested_langs
.extend(current_langs
)
2689 requested_langs
= orderedSet(requested_langs
)
2690 elif normal_sub_langs
:
2691 requested_langs
= ['en'] if 'en' in normal_sub_langs
else normal_sub_langs
[:1]
2693 requested_langs
= ['en'] if 'en' in all_sub_langs
else all_sub_langs
[:1]
2695 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2697 formats_query
= self
.params
.get('subtitlesformat', 'best')
2698 formats_preference
= formats_query
.split('/') if formats_query
else []
2700 for lang
in requested_langs
:
2701 formats
= available_subs
.get(lang
)
2703 self
.report_warning(f
'{lang} subtitles not available for {video_id}')
2705 for ext
in formats_preference
:
2709 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2715 self
.report_warning(
2716 'No subtitle format found matching "%s" for language %s, '
2717 'using %s' % (formats_query
, lang
, f
['ext']))
2721 def _forceprint(self
, key
, info_dict
):
2722 if info_dict
is None:
2724 info_copy
= info_dict
.copy()
2725 info_copy
['formats_table'] = self
.render_formats_table(info_dict
)
2726 info_copy
['thumbnails_table'] = self
.render_thumbnails_table(info_dict
)
2727 info_copy
['subtitles_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('subtitles'))
2728 info_copy
['automatic_captions_table'] = self
.render_subtitles_table(info_dict
.get('id'), info_dict
.get('automatic_captions'))
2730 def format_tmpl(tmpl
):
2731 mobj
= re
.match(r
'\w+(=?)$', tmpl
)
2732 if mobj
and mobj
.group(1):
2733 return f
'{tmpl[:-1]} = %({tmpl[:-1]})r'
2735 return f
'%({tmpl})s'
2738 for tmpl
in self
.params
['forceprint'].get(key
, []):
2739 self
.to_stdout(self
.evaluate_outtmpl(format_tmpl(tmpl
), info_copy
))
2741 for tmpl
, file_tmpl
in self
.params
['print_to_file'].get(key
, []):
2742 filename
= self
.prepare_filename(info_dict
, outtmpl
=file_tmpl
)
2743 tmpl
= format_tmpl(tmpl
)
2744 self
.to_screen(f
'[info] Writing {tmpl!r} to: {filename}')
2745 if self
._ensure
_dir
_exists
(filename
):
2746 with open(filename
, 'a', encoding
='utf-8') as f
:
2747 f
.write(self
.evaluate_outtmpl(tmpl
, info_copy
) + '\n')
2749 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2750 def print_mandatory(field
, actual_field
=None):
2751 if actual_field
is None:
2752 actual_field
= field
2753 if (self
.params
.get('force%s' % field
, False)
2754 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2755 self
.to_stdout(info_dict
[actual_field
])
2757 def print_optional(field
):
2758 if (self
.params
.get('force%s' % field
, False)
2759 and info_dict
.get(field
) is not None):
2760 self
.to_stdout(info_dict
[field
])
2762 info_dict
= info_dict
.copy()
2763 if filename
is not None:
2764 info_dict
['filename'] = filename
2765 if info_dict
.get('requested_formats') is not None:
2766 # For RTMP URLs, also include the playpath
2767 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2768 elif info_dict
.get('url'):
2769 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2771 if (self
.params
.get('forcejson')
2772 or self
.params
['forceprint'].get('video')
2773 or self
.params
['print_to_file'].get('video')):
2774 self
.post_extract(info_dict
)
2775 self
._forceprint
('video', info_dict
)
2777 print_mandatory('title')
2778 print_mandatory('id')
2779 print_mandatory('url', 'urls')
2780 print_optional('thumbnail')
2781 print_optional('description')
2782 print_optional('filename')
2783 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2784 self
.to_stdout(formatSeconds(info_dict
['duration']))
2785 print_mandatory('format')
2787 if self
.params
.get('forcejson'):
2788 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2790 def dl(self
, name
, info
, subtitle
=False, test
=False):
2791 if not info
.get('url'):
2792 self
.raise_no_formats(info
, True)
2795 verbose
= self
.params
.get('verbose')
2798 'quiet': self
.params
.get('quiet') or not verbose
,
2800 'noprogress': not verbose
,
2802 'skip_unavailable_fragments': False,
2803 'keep_fragments': False,
2805 '_no_ytdl_file': True,
2808 params
= self
.params
2809 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2811 for ph
in self
._progress
_hooks
:
2812 fd
.add_progress_hook(ph
)
2814 (f
['url'].split(',')[0] + ',<data>' if f
['url'].startswith('data:') else f
['url'])
2815 for f
in info
.get('requested_formats', []) or [info
])
2816 self
.write_debug('Invoking downloader on "%s"' % urls
)
2818 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2819 # But it may contain objects that are not deep-copyable
2820 new_info
= self
._copy
_infodict
(info
)
2821 if new_info
.get('http_headers') is None:
2822 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2823 return fd
.download(name
, new_info
, subtitle
)
2825 def existing_file(self
, filepaths
, *, default_overwrite
=True):
2826 existing_files
= list(filter(os
.path
.exists
, orderedSet(filepaths
)))
2827 if existing_files
and not self
.params
.get('overwrites', default_overwrite
):
2828 return existing_files
[0]
2830 for file in existing_files
:
2831 self
.report_file_delete(file)
2835 def process_info(self
, info_dict
):
2836 """Process a single resolved IE result. (Modifies it in-place)"""
2838 assert info_dict
.get('_type', 'video') == 'video'
2839 original_infodict
= info_dict
2841 if 'format' not in info_dict
and 'ext' in info_dict
:
2842 info_dict
['format'] = info_dict
['ext']
2844 # This is mostly just for backward compatibility of process_info
2845 # As a side-effect, this allows for format-specific filters
2846 if self
._match
_entry
(info_dict
) is not None:
2847 info_dict
['__write_download_archive'] = 'ignore'
2850 # Does nothing under normal operation - for backward compatibility of process_info
2851 self
.post_extract(info_dict
)
2852 self
._num
_downloads
+= 1
2854 # info_dict['_filename'] needs to be set for backward compatibility
2855 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2856 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2860 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2862 if self
.params
.get('simulate'):
2863 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2866 if full_filename
is None:
2868 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2870 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2873 if self
._write
_description
('video', info_dict
,
2874 self
.prepare_filename(info_dict
, 'description')) is None:
2877 sub_files
= self
._write
_subtitles
(info_dict
, temp_filename
)
2878 if sub_files
is None:
2880 files_to_move
.update(dict(sub_files
))
2882 thumb_files
= self
._write
_thumbnails
(
2883 'video', info_dict
, temp_filename
, self
.prepare_filename(info_dict
, 'thumbnail'))
2884 if thumb_files
is None:
2886 files_to_move
.update(dict(thumb_files
))
2888 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2889 _infojson_written
= self
._write
_info
_json
('video', info_dict
, infofn
)
2890 if _infojson_written
:
2891 info_dict
['infojson_filename'] = infofn
2892 # For backward compatibility, even though it was a private field
2893 info_dict
['__infojson_filename'] = infofn
2894 elif _infojson_written
is None:
2897 # Note: Annotations are deprecated
2899 if self
.params
.get('writeannotations', False):
2900 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2902 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2904 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2905 self
.to_screen('[info] Video annotations are already present')
2906 elif not info_dict
.get('annotations'):
2907 self
.report_warning('There are no annotations to write.')
2910 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2911 with open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2912 annofile
.write(info_dict
['annotations'])
2913 except (KeyError, TypeError):
2914 self
.report_warning('There are no annotations to write.')
2916 self
.report_error('Cannot write annotations file: ' + annofn
)
2919 # Write internet shortcut files
2920 def _write_link_file(link_type
):
2921 url
= try_get(info_dict
['webpage_url'], iri_to_uri
)
2923 self
.report_warning(
2924 f
'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2926 linkfn
= replace_extension(self
.prepare_filename(info_dict
, 'link'), link_type
, info_dict
.get('ext'))
2927 if not self
._ensure
_dir
_exists
(encodeFilename(linkfn
)):
2929 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2930 self
.to_screen(f
'[info] Internet shortcut (.{link_type}) is already present')
2933 self
.to_screen(f
'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2934 with open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8',
2935 newline
='\r\n' if link_type
== 'url' else '\n') as linkfile
:
2936 template_vars
= {'url': url}
2937 if link_type
== 'desktop':
2938 template_vars
['filename'] = linkfn
[:-(len(link_type
) + 1)]
2939 linkfile
.write(LINK_TEMPLATES
[link_type
] % template_vars
)
2941 self
.report_error(f
'Cannot write internet shortcut {linkfn}')
2946 'url': self
.params
.get('writeurllink'),
2947 'webloc': self
.params
.get('writewebloclink'),
2948 'desktop': self
.params
.get('writedesktoplink'),
2950 if self
.params
.get('writelink'):
2951 link_type
= ('webloc' if sys
.platform
== 'darwin'
2952 else 'desktop' if sys
.platform
.startswith('linux')
2954 write_links
[link_type
] = True
2956 if any(should_write
and not _write_link_file(link_type
)
2957 for link_type
, should_write
in write_links
.items()):
2960 def replace_info_dict(new_info
):
2962 if new_info
== info_dict
:
2965 info_dict
.update(new_info
)
2968 new_info
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2969 replace_info_dict(new_info
)
2970 except PostProcessingError
as err
:
2971 self
.report_error('Preprocessing: %s' % str(err
))
2974 if self
.params
.get('skip_download'):
2975 info_dict
['filepath'] = temp_filename
2976 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2977 info_dict
['__files_to_move'] = files_to_move
2978 replace_info_dict(self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
))
2979 info_dict
['__write_download_archive'] = self
.params
.get('force_write_download_archive')
2982 info_dict
.setdefault('__postprocessors', [])
2985 def existing_video_file(*filepaths
):
2986 ext
= info_dict
.get('ext')
2987 converted
= lambda file: replace_extension(file, self
.params
.get('final_ext') or ext
, ext
)
2988 file = self
.existing_file(itertools
.chain(*zip(map(converted
, filepaths
), filepaths
)),
2989 default_overwrite
=False)
2991 info_dict
['ext'] = os
.path
.splitext(file)[1][1:]
2995 if info_dict
.get('requested_formats') is not None:
2997 def compatible_formats(formats
):
2998 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2999 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
3000 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
3001 if len(video_formats
) > 2 or len(audio_formats
) > 2:
3005 exts
= {format.get('ext') for format in formats}
3007 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'}
,
3010 for ext_sets
in COMPATIBLE_EXTS
:
3011 if ext_sets
.issuperset(exts
):
3013 # TODO: Check acodec/vcodec
3016 requested_formats
= info_dict
['requested_formats']
3017 old_ext
= info_dict
['ext']
3018 if self
.params
.get('merge_output_format') is None:
3019 if not compatible_formats(requested_formats
):
3020 info_dict
['ext'] = 'mkv'
3021 self
.report_warning(
3022 'Requested formats are incompatible for merge and will be merged into mkv')
3023 if (info_dict
['ext'] == 'webm'
3024 and info_dict
.get('thumbnails')
3025 # check with type instead of pp_key, __name__, or isinstance
3026 # since we dont want any custom PPs to trigger this
3027 and any(type(pp
) == EmbedThumbnailPP
for pp
in self
._pps
['post_process'])):
3028 info_dict
['ext'] = 'mkv'
3029 self
.report_warning(
3030 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3031 new_ext
= info_dict
['ext']
3033 def correct_ext(filename
, ext
=new_ext
):
3036 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
3038 os
.path
.splitext(filename
)[0]
3039 if filename_real_ext
in (old_ext
, new_ext
)
3041 return f
'{filename_wo_ext}.{ext}'
3043 # Ensure filename always has a correct extension for successful merge
3044 full_filename
= correct_ext(full_filename
)
3045 temp_filename
= correct_ext(temp_filename
)
3046 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3047 info_dict
['__real_download'] = False
3050 merger
= FFmpegMergerPP(self
)
3052 fd
= get_suitable_downloader(info_dict
, self
.params
, to_stdout
=temp_filename
== '-')
3053 if dl_filename
is not None:
3054 self
.report_file_already_downloaded(dl_filename
)
3056 for f
in requested_formats
if fd
!= FFmpegFD
else []:
3057 f
['filepath'] = fname
= prepend_extension(
3058 correct_ext(temp_filename
, info_dict
['ext']),
3059 'f%s' % f
['format_id'], info_dict
['ext'])
3060 downloaded
.append(fname
)
3061 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
3062 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3063 info_dict
['__real_download'] = real_download
3065 if self
.params
.get('allow_unplayable_formats'):
3066 self
.report_warning(
3067 'You have requested merging of multiple formats '
3068 'while also allowing unplayable formats to be downloaded. '
3069 'The formats won\'t be merged to prevent data corruption.')
3070 elif not merger
.available
:
3071 msg
= 'You have requested merging of multiple formats but ffmpeg is not installed'
3072 if not self
.params
.get('ignoreerrors'):
3073 self
.report_error(f
'{msg}. Aborting due to --abort-on-error')
3075 self
.report_warning(f
'{msg}. The formats won\'t be merged')
3077 if temp_filename
== '-':
3078 reason
= ('using a downloader other than ffmpeg' if FFmpegFD
.can_merge_formats(info_dict
, self
.params
)
3079 else 'but the formats are incompatible for simultaneous download' if merger
.available
3080 else 'but ffmpeg is not installed')
3081 self
.report_warning(
3082 f
'You have requested downloading multiple formats to stdout {reason}. '
3083 'The formats will be streamed one after the other')
3084 fname
= temp_filename
3085 for f
in requested_formats
:
3086 new_info
= dict(info_dict
)
3087 del new_info
['requested_formats']
3089 if temp_filename
!= '-':
3090 fname
= prepend_extension(
3091 correct_ext(temp_filename
, new_info
['ext']),
3092 'f%s' % f
['format_id'], new_info
['ext'])
3093 if not self
._ensure
_dir
_exists
(fname
):
3095 f
['filepath'] = fname
3096 downloaded
.append(fname
)
3097 partial_success
, real_download
= self
.dl(fname
, new_info
)
3098 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
3099 success
= success
and partial_success
3101 if downloaded
and merger
.available
and not self
.params
.get('allow_unplayable_formats'):
3102 info_dict
['__postprocessors'].append(merger
)
3103 info_dict
['__files_to_merge'] = downloaded
3104 # Even if there were no downloads, it is being merged only now
3105 info_dict
['__real_download'] = True
3107 for file in downloaded
:
3108 files_to_move
[file] = None
3110 # Just a single file
3111 dl_filename
= existing_video_file(full_filename
, temp_filename
)
3112 if dl_filename
is None or dl_filename
== temp_filename
:
3113 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3114 # So we should try to resume the download
3115 success
, real_download
= self
.dl(temp_filename
, info_dict
)
3116 info_dict
['__real_download'] = real_download
3118 self
.report_file_already_downloaded(dl_filename
)
3120 dl_filename
= dl_filename
or temp_filename
3121 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
3123 except network_exceptions
as err
:
3124 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
3126 except OSError as err
:
3127 raise UnavailableVideoError(err
)
3128 except (ContentTooShortError
, ) as err
:
3129 self
.report_error(f
'content too short (expected {err.expected} bytes and served {err.downloaded})')
3132 if success
and full_filename
!= '-':
3136 fixup_policy
= self
.params
.get('fixup')
3137 vid
= info_dict
['id']
3139 if fixup_policy
in ('ignore', 'never'):
3141 elif fixup_policy
== 'warn':
3143 elif fixup_policy
!= 'force':
3144 assert fixup_policy
in ('detect_or_warn', None)
3145 if not info_dict
.get('__real_download'):
3148 def ffmpeg_fixup(cndn
, msg
, cls
):
3152 self
.report_warning(f
'{vid}: {msg}')
3156 info_dict
['__postprocessors'].append(pp
)
3158 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
3160 stretched_ratio
= info_dict
.get('stretched_ratio')
3162 stretched_ratio
not in (1, None),
3163 f
'Non-uniform pixel ratio {stretched_ratio}',
3164 FFmpegFixupStretchedPP
)
3167 (info_dict
.get('requested_formats') is None
3168 and info_dict
.get('container') == 'm4a_dash'
3169 and info_dict
.get('ext') == 'm4a'),
3170 'writing DASH m4a. Only some players support this container',
3173 downloader
= get_suitable_downloader(info_dict
, self
.params
) if 'protocol' in info_dict
else None
3174 downloader
= downloader
.__name
__ if downloader
else None
3176 if info_dict
.get('requested_formats') is None: # Not necessary if doing merger
3177 ffmpeg_fixup(downloader
== 'HlsFD',
3178 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3180 ffmpeg_fixup(info_dict
.get('is_live') and downloader
== 'DashSegmentsFD',
3181 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP
)
3183 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP
)
3184 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP
)
3188 replace_info_dict(self
.post_process(dl_filename
, info_dict
, files_to_move
))
3189 except PostProcessingError
as err
:
3190 self
.report_error('Postprocessing: %s' % str(err
))
3193 for ph
in self
._post
_hooks
:
3194 ph(info_dict
['filepath'])
3195 except Exception as err
:
3196 self
.report_error('post hooks: %s' % str(err
))
3198 info_dict
['__write_download_archive'] = True
3200 if self
.params
.get('force_write_download_archive'):
3201 info_dict
['__write_download_archive'] = True
3203 # Make sure the info_dict was modified in-place
3204 assert info_dict
is original_infodict
3206 max_downloads
= self
.params
.get('max_downloads')
3207 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
3208 raise MaxDownloadsReached()
3210 def __download_wrapper(self
, func
):
3211 @functools.wraps(func
)
3212 def wrapper(*args
, **kwargs
):
3214 res
= func(*args
, **kwargs
)
3215 except UnavailableVideoError
as e
:
3216 self
.report_error(e
)
3217 except MaxDownloadsReached
as e
:
3218 self
.to_screen(f
'[info] {e}')
3220 except DownloadCancelled
as e
:
3221 self
.to_screen(f
'[info] {e}')
3222 if not self
.params
.get('break_per_url'):
3225 if self
.params
.get('dump_single_json', False):
3226 self
.post_extract(res
)
3227 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
3230 def download(self
, url_list
):
3231 """Download a given list of URLs."""
3232 url_list
= variadic(url_list
) # Passing a single URL is a common mistake
3233 outtmpl
= self
.outtmpl_dict
['default']
3234 if (len(url_list
) > 1
3236 and '%' not in outtmpl
3237 and self
.params
.get('max_downloads') != 1):
3238 raise SameFileError(outtmpl
)
3240 for url
in url_list
:
3241 self
.__download
_wrapper
(self
.extract_info
)(
3242 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
3244 return self
._download
_retcode
3246 def download_with_info_file(self
, info_filename
):
3247 with contextlib
.closing(fileinput
.FileInput(
3248 [info_filename
], mode
='r',
3249 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
3250 # FileInput doesn't have a read method, we can't call json.load
3251 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
3253 self
.__download
_wrapper
(self
.process_ie_result
)(info
, download
=True)
3254 except (DownloadError
, EntryNotInPlaylist
, ReExtractInfo
) as e
:
3255 if not isinstance(e
, EntryNotInPlaylist
):
3256 self
.to_stderr('\r')
3257 webpage_url
= info
.get('webpage_url')
3258 if webpage_url
is not None:
3259 self
.report_warning(f
'The info failed to download: {e}; trying with URL {webpage_url}')
3260 return self
.download([webpage_url
])
3263 return self
._download
_retcode
3266 def sanitize_info(info_dict
, remove_private_keys
=False):
3267 ''' Sanitize the infodict for converting to json '''
3268 if info_dict
is None:
3270 info_dict
.setdefault('epoch', int(time
.time()))
3271 info_dict
.setdefault('_type', 'video')
3273 if remove_private_keys
:
3274 reject
= lambda k
, v
: v
is None or (k
.startswith('_') and k
!= '_type') or k
in {
3275 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3276 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3279 reject
= lambda k
, v
: False
3282 if isinstance(obj
, dict):
3283 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3284 elif isinstance(obj
, (list, tuple, set, LazyList
)):
3285 return list(map(filter_fn
, obj
))
3286 elif obj
is None or isinstance(obj
, (str, int, float, bool)):
3291 return filter_fn(info_dict
)
3294 def filter_requested_info(info_dict
, actually_filter
=True):
3295 ''' Alias of sanitize_info for backward compatibility '''
3296 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
3299 def post_extract(info_dict
):
3300 def actual_post_extract(info_dict
):
3301 if info_dict
.get('_type') in ('playlist', 'multi_video'):
3302 for video_dict
in info_dict
.get('entries', {}):
3303 actual_post_extract(video_dict
or {})
3306 post_extractor
= info_dict
.pop('__post_extractor', None) or (lambda: {})
3307 info_dict
.update(post_extractor())
3309 actual_post_extract(info_dict
or {})
3311 def run_pp(self
, pp
, infodict
):
3312 files_to_delete
= []
3313 if '__files_to_move' not in infodict
:
3314 infodict
['__files_to_move'] = {}
3316 files_to_delete
, infodict
= pp
.run(infodict
)
3317 except PostProcessingError
as e
:
3318 # Must be True and not 'only_download'
3319 if self
.params
.get('ignoreerrors') is True:
3320 self
.report_error(e
)
3324 if not files_to_delete
:
3326 if self
.params
.get('keepvideo', False):
3327 for f
in files_to_delete
:
3328 infodict
['__files_to_move'].setdefault(f
, '')
3330 for old_filename
in set(files_to_delete
):
3331 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
3333 os
.remove(encodeFilename(old_filename
))
3335 self
.report_warning('Unable to remove downloaded original file')
3336 if old_filename
in infodict
['__files_to_move']:
3337 del infodict
['__files_to_move'][old_filename
]
3340 def run_all_pps(self
, key
, info
, *, additional_pps
=None):
3341 self
._forceprint
(key
, info
)
3342 for pp
in (additional_pps
or []) + self
._pps
[key
]:
3343 info
= self
.run_pp(pp
, info
)
3346 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
3347 info
= dict(ie_info
)
3348 info
['__files_to_move'] = files_to_move
or {}
3349 info
= self
.run_all_pps(key
, info
)
3350 return info
, info
.pop('__files_to_move', None)
3352 def post_process(self
, filename
, info
, files_to_move
=None):
3353 """Run all the postprocessors on the given file."""
3354 info
['filepath'] = filename
3355 info
['__files_to_move'] = files_to_move
or {}
3356 info
= self
.run_all_pps('post_process', info
, additional_pps
=info
.get('__postprocessors'))
3357 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3358 del info
['__files_to_move']
3359 return self
.run_all_pps('after_move', info
)
3361 def _make_archive_id(self
, info_dict
):
3362 video_id
= info_dict
.get('id')
3365 # Future-proof against any change in case
3366 # and backwards compatibility with prior versions
3367 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3368 if extractor
is None:
3369 url
= str_or_none(info_dict
.get('url'))
3372 # Try to find matching extractor for the URL and take its ie_key
3373 for ie_key
, ie
in self
._ies
.items():
3374 if ie
.suitable(url
):
3379 return f
'{extractor.lower()} {video_id}'
3381 def in_download_archive(self
, info_dict
):
3382 fn
= self
.params
.get('download_archive')
3386 vid_id
= self
._make
_archive
_id
(info_dict
)
3388 return False # Incomplete video information
3390 return vid_id
in self
.archive
3392 def record_download_archive(self
, info_dict
):
3393 fn
= self
.params
.get('download_archive')
3396 vid_id
= self
._make
_archive
_id
(info_dict
)
3398 self
.write_debug(f
'Adding to archive: {vid_id}')
3399 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3400 archive_file
.write(vid_id
+ '\n')
3401 self
.archive
.add(vid_id
)
3404 def format_resolution(format
, default
='unknown'):
3405 if format
.get('vcodec') == 'none' and format
.get('acodec') != 'none':
3407 if format
.get('resolution') is not None:
3408 return format
['resolution']
3409 if format
.get('width') and format
.get('height'):
3410 return '%dx%d' % (format
['width'], format
['height'])
3411 elif format
.get('height'):
3412 return '%sp' % format
['height']
3413 elif format
.get('width'):
3414 return '%dx?' % format
['width']
3417 def _list_format_headers(self
, *headers
):
3418 if self
.params
.get('listformats_table', True) is not False:
3419 return [self
._format
_screen
(header
, self
.Styles
.HEADERS
) for header
in headers
]
3422 def _format_note(self
, fdict
):
3424 if fdict
.get('ext') in ['f4f', 'f4m']:
3425 res
+= '(unsupported)'
3426 if fdict
.get('language'):
3429 res
+= '[%s]' % fdict
['language']
3430 if fdict
.get('format_note') is not None:
3433 res
+= fdict
['format_note']
3434 if fdict
.get('tbr') is not None:
3437 res
+= '%4dk' % fdict
['tbr']
3438 if fdict
.get('container') is not None:
3441 res
+= '%s container' % fdict
['container']
3442 if (fdict
.get('vcodec') is not None
3443 and fdict
.get('vcodec') != 'none'):
3446 res
+= fdict
['vcodec']
3447 if fdict
.get('vbr') is not None:
3449 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3451 if fdict
.get('vbr') is not None:
3452 res
+= '%4dk' % fdict
['vbr']
3453 if fdict
.get('fps') is not None:
3456 res
+= '%sfps' % fdict
['fps']
3457 if fdict
.get('acodec') is not None:
3460 if fdict
['acodec'] == 'none':
3463 res
+= '%-5s' % fdict
['acodec']
3464 elif fdict
.get('abr') is not None:
3468 if fdict
.get('abr') is not None:
3469 res
+= '@%3dk' % fdict
['abr']
3470 if fdict
.get('asr') is not None:
3471 res
+= ' (%5dHz)' % fdict
['asr']
3472 if fdict
.get('filesize') is not None:
3475 res
+= format_bytes(fdict
['filesize'])
3476 elif fdict
.get('filesize_approx') is not None:
3479 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3482 def render_formats_table(self
, info_dict
):
3483 if not info_dict
.get('formats') and not info_dict
.get('url'):
3486 formats
= info_dict
.get('formats', [info_dict
])
3487 if not self
.params
.get('listformats_table', True) is not False:
3490 format_field(f
, 'format_id'),
3491 format_field(f
, 'ext'),
3492 self
.format_resolution(f
),
3493 self
._format
_note
(f
)
3494 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3495 return render_table(['format code', 'extension', 'resolution', 'note'], table
, extra_gap
=1)
3497 delim
= self
._format
_screen
('\u2502', self
.Styles
.DELIM
, '|', test_encoding
=True)
3500 self
._format
_screen
(format_field(f
, 'format_id'), self
.Styles
.ID
),
3501 format_field(f
, 'ext'),
3502 format_field(f
, func
=self
.format_resolution
, ignore
=('audio only', 'images')),
3503 format_field(f
, 'fps', '\t%d'),
3504 format_field(f
, 'dynamic_range', '%s', ignore
=(None, 'SDR')).replace('HDR', ''),
3506 format_field(f
, 'filesize', ' \t%s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~\t%s', func
=format_bytes
),
3507 format_field(f
, 'tbr', '\t%dk'),
3508 shorten_protocol_name(f
.get('protocol', '')),
3510 format_field(f
, 'vcodec', default
='unknown').replace(
3511 'none', 'images' if f
.get('acodec') == 'none'
3512 else self
._format
_screen
('audio only', self
.Styles
.SUPPRESS
)),
3513 format_field(f
, 'vbr', '\t%dk'),
3514 format_field(f
, 'acodec', default
='unknown').replace(
3515 'none', '' if f
.get('vcodec') == 'none'
3516 else self
._format
_screen
('video only', self
.Styles
.SUPPRESS
)),
3517 format_field(f
, 'abr', '\t%dk'),
3518 format_field(f
, 'asr', '\t%dHz'),
3520 self
._format
_screen
('UNSUPPORTED', 'light red') if f
.get('ext') in ('f4f', 'f4m') else None,
3521 format_field(f
, 'language', '[%s]'),
3522 join_nonempty(format_field(f
, 'format_note'),
3523 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3526 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3527 header_line
= self
._list
_format
_headers
(
3528 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim
, '\tFILESIZE', '\tTBR', 'PROTO',
3529 delim
, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3531 return render_table(
3532 header_line
, table
, hide_empty
=True,
3533 delim
=self
._format
_screen
('\u2500', self
.Styles
.DELIM
, '-', test_encoding
=True))
3535 def render_thumbnails_table(self
, info_dict
):
3536 thumbnails
= list(info_dict
.get('thumbnails') or [])
3539 return render_table(
3540 self
._list
_format
_headers
('ID', 'Width', 'Height', 'URL'),
3541 [[t
.get('id'), t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
])
3543 def render_subtitles_table(self
, video_id
, subtitles
):
3544 def _row(lang
, formats
):
3545 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3546 if len(set(names
)) == 1:
3547 names
= [] if names
[0] == 'unknown' else names
[:1]
3548 return [lang
, ', '.join(names
), ', '.join(exts
)]
3552 return render_table(
3553 self
._list
_format
_headers
('Language', 'Name', 'Formats'),
3554 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3557 def __list_table(self
, video_id
, name
, func
, *args
):
3560 self
.to_screen(f
'{video_id} has no {name}')
3562 self
.to_screen(f
'[info] Available {name} for {video_id}:')
3563 self
.to_stdout(table
)
3565 def list_formats(self
, info_dict
):
3566 self
.__list
_table
(info_dict
['id'], 'formats', self
.render_formats_table
, info_dict
)
3568 def list_thumbnails(self
, info_dict
):
3569 self
.__list
_table
(info_dict
['id'], 'thumbnails', self
.render_thumbnails_table
, info_dict
)
3571 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3572 self
.__list
_table
(video_id
, name
, self
.render_subtitles_table
, video_id
, subtitles
)
3574 def urlopen(self
, req
):
3575 """ Start an HTTP download """
3576 if isinstance(req
, str):
3577 req
= sanitized_Request(req
)
3578 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3580 def print_debug_header(self
):
3581 if not self
.params
.get('verbose'):
3584 def get_encoding(stream
):
3585 ret
= str(getattr(stream
, 'encoding', 'missing (%s)' % type(stream
).__name
__))
3586 if not supports_terminal_sequences(stream
):
3587 from .compat
import WINDOWS_VT_MODE
3588 ret
+= ' (No VT)' if WINDOWS_VT_MODE
is False else ' (No ANSI)'
3591 encoding_str
= 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3592 locale
.getpreferredencoding(),
3593 sys
.getfilesystemencoding(),
3594 get_encoding(self
._out
_files
['screen']), get_encoding(self
._out
_files
['error']),
3595 self
.get_encoding())
3597 logger
= self
.params
.get('logger')
3599 write_debug
= lambda msg
: logger
.debug(f
'[debug] {msg}')
3600 write_debug(encoding_str
)
3602 write_string(f
'[debug] {encoding_str}\n', encoding
=None)
3603 write_debug
= lambda msg
: self
._write
_string
(f
'[debug] {msg}\n')
3605 source
= detect_variant()
3606 write_debug(join_nonempty(
3607 'yt-dlp version', __version__
,
3608 f
'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD
else '',
3609 '' if source
== 'unknown' else f
'({source})',
3611 if not _LAZY_LOADER
:
3612 if os
.environ
.get('YTDLP_NO_LAZY_EXTRACTORS'):
3613 write_debug('Lazy loading extractors is forcibly disabled')
3615 write_debug('Lazy loading extractors is disabled')
3616 if plugin_extractors
or plugin_postprocessors
:
3617 write_debug('Plugins: %s' % [
3618 '%s%s' % (klass
.__name
__, '' if klass
.__name
__ == name
else f
' as {name}')
3619 for name
, klass
in itertools
.chain(plugin_extractors
.items(), plugin_postprocessors
.items())])
3620 if self
.params
.get('compat_opts'):
3621 write_debug('Compatibility options: %s' % ', '.join(self
.params
.get('compat_opts')))
3623 if source
== 'source':
3626 ['git', 'rev-parse', '--short', 'HEAD'],
3627 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3628 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3629 out
, err
= sp
.communicate_or_kill()
3630 out
= out
.decode().strip()
3631 if re
.match('[0-9a-f]+', out
):
3632 write_debug('Git HEAD: %s' % out
)
3639 def python_implementation():
3640 impl_name
= platform
.python_implementation()
3641 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3642 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3645 write_debug('Python version %s (%s %s) - %s' % (
3646 platform
.python_version(),
3647 python_implementation(),
3648 platform
.architecture()[0],
3651 exe_versions
, ffmpeg_features
= FFmpegPostProcessor
.get_versions_and_features(self
)
3652 ffmpeg_features
= {key for key, val in ffmpeg_features.items() if val}
3654 exe_versions
['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features
)
3656 exe_versions
['rtmpdump'] = rtmpdump_version()
3657 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3658 exe_str
= ', '.join(
3659 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3661 write_debug('exe versions: %s' % exe_str
)
3663 from .cookies
import SECRETSTORAGE_AVAILABLE
, SQLITE_AVAILABLE
3664 from .downloader
.websocket
import has_websockets
3665 from .postprocessor
.embedthumbnail
import has_mutagen
3667 lib_str
= join_nonempty(
3668 compat_brotli
and compat_brotli
.__name
__,
3669 has_certifi
and 'certifi',
3670 compat_pycrypto_AES
and compat_pycrypto_AES
.__name
__.split('.')[0],
3671 SECRETSTORAGE_AVAILABLE
and 'secretstorage',
3672 has_mutagen
and 'mutagen',
3673 SQLITE_AVAILABLE
and 'sqlite',
3674 has_websockets
and 'websockets',
3675 delim
=', ') or 'none'
3676 write_debug('Optional libraries: %s' % lib_str
)
3678 self
._setup
_opener
()
3680 for handler
in self
._opener
.handlers
:
3681 if hasattr(handler
, 'proxies'):
3682 proxy_map
.update(handler
.proxies
)
3683 write_debug(f
'Proxy map: {proxy_map}')
3686 if False and self
.params
.get('call_home'):
3687 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3688 write_debug('Public IP address: %s' % ipaddr
)
3689 latest_version
= self
.urlopen(
3690 'https://yt-dl.org/latest/version').read().decode('utf-8')
3691 if version_tuple(latest_version
) > version_tuple(__version__
):
3692 self
.report_warning(
3693 'You are using an outdated version (newest version: %s)! '
3694 'See https://yt-dl.org/update if you need help updating.' %
3697 def _setup_opener(self
):
3698 if hasattr(self
, '_opener'):
3700 timeout_val
= self
.params
.get('socket_timeout')
3701 self
._socket
_timeout
= 20 if timeout_val
is None else float(timeout_val
)
3703 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3704 opts_cookiefile
= self
.params
.get('cookiefile')
3705 opts_proxy
= self
.params
.get('proxy')
3707 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3709 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3710 if opts_proxy
is not None:
3711 if opts_proxy
== '':
3714 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3716 proxies
= compat_urllib_request
.getproxies()
3717 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3718 if 'http' in proxies
and 'https' not in proxies
:
3719 proxies
['https'] = proxies
['http']
3720 proxy_handler
= PerRequestProxyHandler(proxies
)
3722 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3723 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3724 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3725 redirect_handler
= YoutubeDLRedirectHandler()
3726 data_handler
= urllib
.request
.DataHandler()
3728 # When passing our own FileHandler instance, build_opener won't add the
3729 # default FileHandler and allows us to disable the file protocol, which
3730 # can be used for malicious purposes (see
3731 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3732 file_handler
= compat_urllib_request
.FileHandler()
3734 def file_open(*args
, **kwargs
):
3735 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3736 file_handler
.file_open
= file_open
3738 opener
= compat_urllib_request
.build_opener(
3739 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3741 # Delete the default user-agent header, which would otherwise apply in
3742 # cases where our custom HTTP handler doesn't come into play
3743 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3744 opener
.addheaders
= []
3745 self
._opener
= opener
3747 def encode(self
, s
):
3748 if isinstance(s
, bytes):
3749 return s
# Already encoded
3752 return s
.encode(self
.get_encoding())
3753 except UnicodeEncodeError as err
:
3754 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3757 def get_encoding(self
):
3758 encoding
= self
.params
.get('encoding')
3759 if encoding
is None:
3760 encoding
= preferredencoding()
3763 def _write_info_json(self
, label
, ie_result
, infofn
, overwrite
=None):
3764 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3765 if overwrite
is None:
3766 overwrite
= self
.params
.get('overwrites', True)
3767 if not self
.params
.get('writeinfojson'):
3770 self
.write_debug(f
'Skipping writing {label} infojson')
3772 elif not self
._ensure
_dir
_exists
(infofn
):
3774 elif not overwrite
and os
.path
.exists(infofn
):
3775 self
.to_screen(f
'[info] {label.title()} metadata is already present')
3778 self
.to_screen(f
'[info] Writing {label} metadata as JSON to: {infofn}')
3780 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
3783 self
.report_error(f
'Cannot write {label} metadata to JSON file {infofn}')
3786 def _write_description(self
, label
, ie_result
, descfn
):
3787 ''' Write description and returns True = written, False = skip, None = error '''
3788 if not self
.params
.get('writedescription'):
3791 self
.write_debug(f
'Skipping writing {label} description')
3793 elif not self
._ensure
_dir
_exists
(descfn
):
3795 elif not self
.params
.get('overwrites', True) and os
.path
.exists(descfn
):
3796 self
.to_screen(f
'[info] {label.title()} description is already present')
3797 elif ie_result
.get('description') is None:
3798 self
.report_warning(f
'There\'s no {label} description to write')
3802 self
.to_screen(f
'[info] Writing {label} description to: {descfn}')
3803 with open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
3804 descfile
.write(ie_result
['description'])
3806 self
.report_error(f
'Cannot write {label} description file {descfn}')
3810 def _write_subtitles(self
, info_dict
, filename
):
3811 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3813 subtitles
= info_dict
.get('requested_subtitles')
3814 if not subtitles
or not (self
.params
.get('writesubtitles') or self
.params
.get('writeautomaticsub')):
3815 # subtitles download errors are already managed as troubles in relevant IE
3816 # that way it will silently go on when used with unsupporting IE
3819 sub_filename_base
= self
.prepare_filename(info_dict
, 'subtitle')
3820 if not sub_filename_base
:
3821 self
.to_screen('[info] Skipping writing video subtitles')
3823 for sub_lang
, sub_info
in subtitles
.items():
3824 sub_format
= sub_info
['ext']
3825 sub_filename
= subtitles_filename(filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
3826 sub_filename_final
= subtitles_filename(sub_filename_base
, sub_lang
, sub_format
, info_dict
.get('ext'))
3827 existing_sub
= self
.existing_file((sub_filename_final
, sub_filename
))
3829 self
.to_screen(f
'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3830 sub_info
['filepath'] = existing_sub
3831 ret
.append((existing_sub
, sub_filename_final
))
3834 self
.to_screen(f
'[info] Writing video subtitles to: {sub_filename}')
3835 if sub_info
.get('data') is not None:
3837 # Use newline='' to prevent conversion of newline characters
3838 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3839 with open(sub_filename
, 'w', encoding
='utf-8', newline
='') as subfile
:
3840 subfile
.write(sub_info
['data'])
3841 sub_info
['filepath'] = sub_filename
3842 ret
.append((sub_filename
, sub_filename_final
))
3845 self
.report_error(f
'Cannot write video subtitles file {sub_filename}')
3849 sub_copy
= sub_info
.copy()
3850 sub_copy
.setdefault('http_headers', info_dict
.get('http_headers'))
3851 self
.dl(sub_filename
, sub_copy
, subtitle
=True)
3852 sub_info
['filepath'] = sub_filename
3853 ret
.append((sub_filename
, sub_filename_final
))
3854 except (DownloadError
, ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
3855 msg
= f
'Unable to download video subtitles for {sub_lang!r}: {err}'
3856 if self
.params
.get('ignoreerrors') is not True: # False or 'only_download'
3857 if not self
.params
.get('ignoreerrors'):
3858 self
.report_error(msg
)
3859 raise DownloadError(msg
)
3860 self
.report_warning(msg
)
3863 def _write_thumbnails(self
, label
, info_dict
, filename
, thumb_filename_base
=None):
3864 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3865 write_all
= self
.params
.get('write_all_thumbnails', False)
3866 thumbnails
, ret
= [], []
3867 if write_all
or self
.params
.get('writethumbnail', False):
3868 thumbnails
= info_dict
.get('thumbnails') or []
3869 multiple
= write_all
and len(thumbnails
) > 1
3871 if thumb_filename_base
is None:
3872 thumb_filename_base
= filename
3873 if thumbnails
and not thumb_filename_base
:
3874 self
.write_debug(f
'Skipping writing {label} thumbnail')
3877 for idx
, t
in list(enumerate(thumbnails
))[::-1]:
3878 thumb_ext
= (f
'{t["id"]}.' if multiple
else '') + determine_ext(t
['url'], 'jpg')
3879 thumb_display_id
= f
'{label} thumbnail {t["id"]}'
3880 thumb_filename
= replace_extension(filename
, thumb_ext
, info_dict
.get('ext'))
3881 thumb_filename_final
= replace_extension(thumb_filename_base
, thumb_ext
, info_dict
.get('ext'))
3883 existing_thumb
= self
.existing_file((thumb_filename_final
, thumb_filename
))
3885 self
.to_screen('[info] %s is already present' % (
3886 thumb_display_id
if multiple
else f
'{label} thumbnail').capitalize())
3887 t
['filepath'] = existing_thumb
3888 ret
.append((existing_thumb
, thumb_filename_final
))
3890 self
.to_screen(f
'[info] Downloading {thumb_display_id} ...')
3892 uf
= self
.urlopen(sanitized_Request(t
['url'], headers
=t
.get('http_headers', {})))
3893 self
.to_screen(f
'[info] Writing {thumb_display_id} to: {thumb_filename}')
3894 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3895 shutil
.copyfileobj(uf
, thumbf
)
3896 ret
.append((thumb_filename
, thumb_filename_final
))
3897 t
['filepath'] = thumb_filename
3898 except network_exceptions
as err
:
3900 self
.report_warning(f
'Unable to download {thumb_display_id}: {err}')
3901 if ret
and not write_all
: