4 from __future__
import absolute_import
, unicode_literals
29 from string
import ascii_letters
30 from zipimport
import zipimporter
34 compat_get_terminal_size
,
40 compat_tokenize_tokenize
,
42 compat_urllib_request
,
43 compat_urllib_request_DataHandler
,
45 from .cookies
import load_cookies
55 DOT_DESKTOP_LINK_TEMPLATE
,
56 DOT_URL_LINK_TEMPLATE
,
57 DOT_WEBLOC_LINK_TEMPLATE
,
87 PerRequestProxyHandler
,
92 process_communicate_or_kill
,
93 register_socks_protocols
,
110 UnavailableVideoError
,
116 YoutubeDLCookieProcessor
,
118 YoutubeDLRedirectHandler
,
120 from .cache
import Cache
121 from .extractor
import (
122 gen_extractor_classes
,
127 from .extractor
.openload
import PhantomJSwrapper
128 from .downloader
import (
130 get_suitable_downloader
,
131 shorten_protocol_name
133 from .downloader
.rtmp
import rtmpdump_version
134 from .postprocessor
import (
136 FFmpegFixupDurationPP
,
139 FFmpegFixupStretchedPP
,
140 FFmpegFixupTimestampPP
,
143 MoveFilesAfterDownloadPP
,
145 from .version
import __version__
147 if compat_os_name
== 'nt':
151 class YoutubeDL(object):
154 YoutubeDL objects are the ones responsible of downloading the
155 actual video file and writing it to disk if the user has requested
156 it, among some other tasks. In most cases there should be one per
157 program. As, given a video URL, the downloader doesn't know how to
158 extract all the needed information, task that InfoExtractors do, it
159 has to pass the URL to one of them.
161 For this, YoutubeDL objects have a method that allows
162 InfoExtractors to be registered in a given order. When it is passed
163 a URL, the YoutubeDL object handles it to the first InfoExtractor it
164 finds that reports being able to handle it. The InfoExtractor extracts
165 all the information about the video or videos the URL refers to, and
166 YoutubeDL process the extracted information, possibly using a File
167 Downloader to download the video.
169 YoutubeDL objects accept a lot of parameters. In order not to saturate
170 the object constructor with arguments, it receives a dictionary of
171 options instead. These options are available through the params
172 attribute for the InfoExtractors to use. The YoutubeDL also
173 registers itself as the downloader in charge for the InfoExtractors
174 that are added to it, so this is a "mutual registration".
178 username: Username for authentication purposes.
179 password: Password for authentication purposes.
180 videopassword: Password for accessing a video.
181 ap_mso: Adobe Pass multiple-system operator identifier.
182 ap_username: Multiple-system operator account username.
183 ap_password: Multiple-system operator account password.
184 usenetrc: Use netrc for authentication instead.
185 verbose: Print additional info to stdout.
186 quiet: Do not print messages to stdout.
187 no_warnings: Do not print out anything for warnings.
188 forceprint: A list of templates to force print
189 forceurl: Force printing final URL. (Deprecated)
190 forcetitle: Force printing title. (Deprecated)
191 forceid: Force printing ID. (Deprecated)
192 forcethumbnail: Force printing thumbnail URL. (Deprecated)
193 forcedescription: Force printing description. (Deprecated)
194 forcefilename: Force printing final filename. (Deprecated)
195 forceduration: Force printing duration. (Deprecated)
196 forcejson: Force printing info_dict as JSON.
197 dump_single_json: Force printing the info_dict of the whole playlist
198 (or video) as a single JSON line.
199 force_write_download_archive: Force writing download archive regardless
200 of 'skip_download' or 'simulate'.
201 simulate: Do not download the video files. If unset (or None),
202 simulate only if listsubtitles, listformats or list_thumbnails is used
203 format: Video format code. see "FORMAT SELECTION" for more details.
204 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
205 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
206 extracting metadata even if the video is not actually
207 available for download (experimental)
208 format_sort: How to sort the video formats. see "Sorting Formats"
210 format_sort_force: Force the given format_sort. see "Sorting Formats"
212 allow_multiple_video_streams: Allow multiple video streams to be merged
214 allow_multiple_audio_streams: Allow multiple audio streams to be merged
216 check_formats Whether to test if the formats are downloadable.
217 Can be True (check all), False (check none)
218 or None (check only if requested by extractor)
219 paths: Dictionary of output paths. The allowed keys are 'home'
220 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
221 outtmpl: Dictionary of templates for output names. Allowed keys
222 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
223 For compatibility with youtube-dl, a single string can also be used
224 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
225 restrictfilenames: Do not allow "&" and spaces in file names
226 trim_file_name: Limit length of filename (extension excluded)
227 windowsfilenames: Force the filenames to be windows compatible
228 ignoreerrors: Do not stop on download errors
229 (Default True when running yt-dlp,
230 but False when directly accessing YoutubeDL class)
231 skip_playlist_after_errors: Number of allowed failures until the rest of
232 the playlist is skipped
233 force_generic_extractor: Force downloader to use the generic extractor
234 overwrites: Overwrite all video and metadata files if True,
235 overwrite only non-video files if None
236 and don't overwrite any file if False
237 For compatibility with youtube-dl,
238 "nooverwrites" may also be used instead
239 playliststart: Playlist item to start at.
240 playlistend: Playlist item to end at.
241 playlist_items: Specific indices of playlist to download.
242 playlistreverse: Download playlist items in reverse order.
243 playlistrandom: Download playlist items in random order.
244 matchtitle: Download only matching titles.
245 rejecttitle: Reject downloads for matching titles.
246 logger: Log messages to a logging.Logger instance.
247 logtostderr: Log messages to stderr instead of stdout.
248 writedescription: Write the video description to a .description file
249 writeinfojson: Write the video description to a .info.json file
250 clean_infojson: Remove private fields from the infojson
251 getcomments: Extract video comments. This will not be written to disk
252 unless writeinfojson is also given
253 writeannotations: Write the video annotations to a .annotations.xml file
254 writethumbnail: Write the thumbnail image to a file
255 allow_playlist_files: Whether to write playlists' description, infojson etc
256 also to disk when using the 'write*' options
257 write_all_thumbnails: Write all thumbnail formats to files
258 writelink: Write an internet shortcut file, depending on the
259 current platform (.url/.webloc/.desktop)
260 writeurllink: Write a Windows internet shortcut file (.url)
261 writewebloclink: Write a macOS internet shortcut file (.webloc)
262 writedesktoplink: Write a Linux internet shortcut file (.desktop)
263 writesubtitles: Write the video subtitles to a file
264 writeautomaticsub: Write the automatically generated subtitles to a file
265 allsubtitles: Deprecated - Use subtitleslangs = ['all']
266 Downloads all the subtitles of the video
267 (requires writesubtitles or writeautomaticsub)
268 listsubtitles: Lists all available subtitles for the video
269 subtitlesformat: The format code for subtitles
270 subtitleslangs: List of languages of the subtitles to download (can be regex).
271 The list may contain "all" to refer to all the available
272 subtitles. The language can be prefixed with a "-" to
273 exclude it from the requested languages. Eg: ['all', '-live_chat']
274 keepvideo: Keep the video file after post-processing
275 daterange: A DateRange object, download only if the upload_date is in the range.
276 skip_download: Skip the actual download of the video file
277 cachedir: Location of the cache files in the filesystem.
278 False to disable filesystem cache.
279 noplaylist: Download single video instead of a playlist if in doubt.
280 age_limit: An integer representing the user's age in years.
281 Unsuitable videos for the given age are skipped.
282 min_views: An integer representing the minimum view count the video
283 must have in order to not be skipped.
284 Videos without view count information are always
285 downloaded. None for no limit.
286 max_views: An integer representing the maximum view count.
287 Videos that are more popular than that are not
289 Videos without view count information are always
290 downloaded. None for no limit.
291 download_archive: File name of a file where all downloads are recorded.
292 Videos already present in the file are not downloaded
294 break_on_existing: Stop the download process after attempting to download a
295 file that is in the archive.
296 break_on_reject: Stop the download process when encountering a video that
297 has been filtered out.
298 cookiefile: File name where cookies should be read from and dumped to
299 cookiesfrombrowser: A tuple containing the name of the browser and the profile
300 name/path from where cookies are loaded.
301 Eg: ('chrome', ) or (vivaldi, 'default')
302 nocheckcertificate:Do not verify SSL certificates
303 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
304 At the moment, this is only supported by YouTube.
305 proxy: URL of the proxy server to use
306 geo_verification_proxy: URL of the proxy to use for IP address verification
307 on geo-restricted sites.
308 socket_timeout: Time to wait for unresponsive hosts, in seconds
309 bidi_workaround: Work around buggy terminals without bidirectional text
310 support, using fridibi
311 debug_printtraffic:Print out sent and received HTTP traffic
312 include_ads: Download ads as well
313 default_search: Prepend this string if an input url is not valid.
314 'auto' for elaborate guessing
315 encoding: Use this encoding instead of the system-specified.
316 extract_flat: Do not resolve URLs, return the immediate result.
317 Pass in 'in_playlist' to only show this behavior for
319 postprocessors: A list of dictionaries, each with an entry
320 * key: The name of the postprocessor. See
321 yt_dlp/postprocessor/__init__.py for a list.
322 * when: When to run the postprocessor. Can be one of
323 pre_process|before_dl|post_process|after_move.
324 Assumed to be 'post_process' if not given
325 post_hooks: A list of functions that get called as the final step
326 for each video file, after all postprocessors have been
327 called. The filename will be passed as the only argument.
328 progress_hooks: A list of functions that get called on download
329 progress, with a dictionary with the entries
330 * status: One of "downloading", "error", or "finished".
331 Check this first and ignore unknown values.
332 * info_dict: The extracted info_dict
334 If status is one of "downloading", or "finished", the
335 following properties may also be present:
336 * filename: The final filename (always present)
337 * tmpfilename: The filename we're currently writing to
338 * downloaded_bytes: Bytes on disk
339 * total_bytes: Size of the whole file, None if unknown
340 * total_bytes_estimate: Guess of the eventual file size,
342 * elapsed: The number of seconds since download started.
343 * eta: The estimated time in seconds, None if unknown
344 * speed: The download speed in bytes/second, None if
346 * fragment_index: The counter of the currently
347 downloaded video fragment.
348 * fragment_count: The number of fragments (= individual
349 files that will be merged)
351 Progress hooks are guaranteed to be called at least once
352 (with status "finished") if the download is successful.
353 merge_output_format: Extension to use when merging formats.
354 final_ext: Expected final extension; used to detect when the file was
355 already downloaded and converted. "merge_output_format" is
356 replaced by this extension when given
357 fixup: Automatically correct known faults of the file.
359 - "never": do nothing
360 - "warn": only emit a warning
361 - "detect_or_warn": check whether we can do anything
362 about it, warn otherwise (default)
363 source_address: Client-side IP address to bind to.
364 call_home: Boolean, true iff we are allowed to contact the
365 yt-dlp servers for debugging. (BROKEN)
366 sleep_interval_requests: Number of seconds to sleep between requests
368 sleep_interval: Number of seconds to sleep before each download when
369 used alone or a lower bound of a range for randomized
370 sleep before each download (minimum possible number
371 of seconds to sleep) when used along with
373 max_sleep_interval:Upper bound of a range for randomized sleep before each
374 download (maximum possible number of seconds to sleep).
375 Must only be used along with sleep_interval.
376 Actual sleep time will be a random float from range
377 [sleep_interval; max_sleep_interval].
378 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
379 listformats: Print an overview of available video formats and exit.
380 list_thumbnails: Print a table of all thumbnails and exit.
381 match_filter: A function that gets called with the info_dict of
383 If it returns a message, the video is ignored.
384 If it returns None, the video is downloaded.
385 match_filter_func in utils.py is one example for this.
386 no_color: Do not emit color codes in output.
387 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
390 Two-letter ISO 3166-2 country code that will be used for
391 explicit geographic restriction bypassing via faking
392 X-Forwarded-For HTTP header
394 IP range in CIDR notation that will be used similarly to
397 The following options determine which downloader is picked:
398 external_downloader: A dictionary of protocol keys and the executable of the
399 external downloader to use for it. The allowed protocols
400 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
401 Set the value to 'native' to use the native downloader
402 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
403 or {'m3u8': 'ffmpeg'} instead.
404 Use the native HLS downloader instead of ffmpeg/avconv
405 if True, otherwise use ffmpeg/avconv if False, otherwise
406 use downloader suggested by extractor if None.
407 compat_opts: Compatibility options. See "Differences in default behavior".
408 The following options do not work when used through the API:
409 filename, abort-on-error, multistreams, no-live-chat,
410 no-clean-infojson, no-playlist-metafiles, no-keep-subs.
411 Refer __init__.py for their implementation
413 The following parameters are not used by YoutubeDL itself, they are used by
414 the downloader (see yt_dlp/downloader/common.py):
415 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
416 max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle,
417 xattr_set_filesize, external_downloader_args, hls_use_mpegts, http_chunk_size.
419 The following options are used by the post processors:
420 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
421 otherwise prefer ffmpeg. (avconv support is deprecated)
422 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
423 to the binary or its containing directory.
424 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
425 and a list of additional command-line arguments for the
426 postprocessor/executable. The dict can also have "PP+EXE" keys
427 which are used when the given exe is used by the given PP.
428 Use 'default' as the name for arguments to passed to all PP
429 For compatibility with youtube-dl, a single list of args
432 The following options are used by the extractors:
433 extractor_retries: Number of times to retry for known errors
434 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
435 hls_split_discontinuity: Split HLS playlists to different formats at
436 discontinuities such as ad breaks (default: False)
437 extractor_args: A dictionary of arguments to be passed to the extractors.
438 See "EXTRACTOR ARGUMENTS" for details.
439 Eg: {'youtube': {'skip': ['dash', 'hls']}}
440 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
441 If True (default), DASH manifests and related
442 data will be downloaded and processed by extractor.
443 You can reduce network I/O by disabling it if you don't
444 care about DASH. (only for youtube)
445 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
446 If True (default), HLS manifests and related
447 data will be downloaded and processed by extractor.
448 You can reduce network I/O by disabling it if you don't
449 care about HLS. (only for youtube)
452 _NUMERIC_FIELDS
= set((
453 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
454 'timestamp', 'upload_year', 'upload_month', 'upload_day',
455 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
456 'average_rating', 'comment_count', 'age_limit',
457 'start_time', 'end_time',
458 'chapter_number', 'season_number', 'episode_number',
459 'track_number', 'disc_number', 'release_year',
465 _pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
466 _printed_messages
= set()
467 _first_webpage_request
= True
468 _download_retcode
= None
469 _num_downloads
= None
471 _playlist_urls
= set()
474 def __init__(self
, params
=None, auto_init
=True):
475 """Create a FileDownloader object with the given options."""
479 self
._ies
_instances
= {}
480 self
._pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
481 self
._printed
_messages
= set()
482 self
._first
_webpage
_request
= True
483 self
._post
_hooks
= []
484 self
._progress
_hooks
= []
485 self
._download
_retcode
= 0
486 self
._num
_downloads
= 0
487 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
488 self
._err
_file
= sys
.stderr
491 'nocheckcertificate': False,
493 self
.params
.update(params
)
494 self
.cache
= Cache(self
)
496 if sys
.version_info
< (3, 6):
498 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
500 if self
.params
.get('allow_unplayable_formats'):
502 'You have asked for unplayable formats to be listed/downloaded. '
503 'This is a developer option intended for debugging. '
504 'If you experience any issues while using this option, DO NOT open a bug report')
506 def check_deprecated(param
, option
, suggestion
):
507 if self
.params
.get(param
) is not None:
508 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
512 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
513 if self
.params
.get('geo_verification_proxy') is None:
514 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
516 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
517 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
518 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
520 for msg
in self
.params
.get('warnings', []):
521 self
.report_warning(msg
)
523 if self
.params
.get('overwrites') is None:
524 self
.params
.pop('overwrites', None)
525 elif self
.params
.get('nooverwrites') is not None:
526 # nooverwrites was unnecessarily changed to overwrites
527 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
528 # This ensures compatibility with both keys
529 self
.params
['overwrites'] = not self
.params
['nooverwrites']
531 self
.params
['nooverwrites'] = not self
.params
['overwrites']
533 if params
.get('bidi_workaround', False):
536 master
, slave
= pty
.openpty()
537 width
= compat_get_terminal_size().columns
541 width_args
= ['-w', str(width
)]
543 stdin
=subprocess
.PIPE
,
545 stderr
=self
._err
_file
)
547 self
._output
_process
= subprocess
.Popen(
548 ['bidiv'] + width_args
, **sp_kwargs
551 self
._output
_process
= subprocess
.Popen(
552 ['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
553 self
._output
_channel
= os
.fdopen(master
, 'rb')
554 except OSError as ose
:
555 if ose
.errno
== errno
.ENOENT
:
556 self
.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.')
560 if (sys
.platform
!= 'win32'
561 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
562 and not params
.get('restrictfilenames', False)):
563 # Unicode filesystem API will throw errors (#1474, #13027)
565 'Assuming --restrict-filenames since file system encoding '
566 'cannot encode all characters. '
567 'Set the LC_ALL environment variable to fix this.')
568 self
.params
['restrictfilenames'] = True
570 self
.outtmpl_dict
= self
.parse_outtmpl()
572 # Creating format selector here allows us to catch syntax errors before the extraction
573 self
.format_selector
= (
574 None if self
.params
.get('format') is None
575 else self
.build_format_selector(self
.params
['format']))
579 """Preload the archive, if any is specified"""
580 def preload_download_archive(fn
):
583 self
.write_debug('Loading archive file %r\n' % fn
)
585 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
586 for line
in archive_file
:
587 self
.archive
.add(line
.strip())
588 except IOError as ioe
:
589 if ioe
.errno
!= errno
.ENOENT
:
595 preload_download_archive(self
.params
.get('download_archive'))
598 self
.print_debug_header()
599 self
.add_default_info_extractors()
601 for pp_def_raw
in self
.params
.get('postprocessors', []):
602 pp_def
= dict(pp_def_raw
)
603 when
= pp_def
.pop('when', 'post_process')
604 pp_class
= get_postprocessor(pp_def
.pop('key'))
605 pp
= pp_class(self
, **compat_kwargs(pp_def
))
606 self
.add_post_processor(pp
, when
=when
)
608 for ph
in self
.params
.get('post_hooks', []):
609 self
.add_post_hook(ph
)
611 for ph
in self
.params
.get('progress_hooks', []):
612 self
.add_progress_hook(ph
)
614 register_socks_protocols()
616 def warn_if_short_id(self
, argv
):
617 # short YouTube ID starting with dash?
619 i
for i
, a
in enumerate(argv
)
620 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
624 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
625 + ['--'] + [argv
[i
] for i
in idxs
]
628 'Long argument string detected. '
629 'Use -- to separate parameters and URLs, like this:\n%s\n' %
630 args_to_str(correct_argv
))
632 def add_info_extractor(self
, ie
):
633 """Add an InfoExtractor object to the end of the list."""
635 if not isinstance(ie
, type):
636 self
._ies
_instances
[ie
.ie_key()] = ie
637 ie
.set_downloader(self
)
639 def get_info_extractor(self
, ie_key
):
641 Get an instance of an IE with name ie_key, it will try to get one from
642 the _ies list, if there's no instance it will create a new one and add
643 it to the extractor list.
645 ie
= self
._ies
_instances
.get(ie_key
)
647 ie
= get_info_extractor(ie_key
)()
648 self
.add_info_extractor(ie
)
651 def add_default_info_extractors(self
):
653 Add the InfoExtractors returned by gen_extractors to the end of the list
655 for ie
in gen_extractor_classes():
656 self
.add_info_extractor(ie
)
658 def add_post_processor(self
, pp
, when
='post_process'):
659 """Add a PostProcessor object to the end of the chain."""
660 self
._pps
[when
].append(pp
)
661 pp
.set_downloader(self
)
663 def add_post_hook(self
, ph
):
664 """Add the post hook"""
665 self
._post
_hooks
.append(ph
)
667 def add_progress_hook(self
, ph
):
668 """Add the progress hook (currently only for the file downloader)"""
669 self
._progress
_hooks
.append(ph
)
671 def _bidi_workaround(self
, message
):
672 if not hasattr(self
, '_output_channel'):
675 assert hasattr(self
, '_output_process')
676 assert isinstance(message
, compat_str
)
677 line_count
= message
.count('\n') + 1
678 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
679 self
._output
_process
.stdin
.flush()
680 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
681 for _
in range(line_count
))
682 return res
[:-len('\n')]
684 def _write_string(self
, message
, out
=None, only_once
=False):
686 if message
in self
._printed
_messages
:
688 self
._printed
_messages
.add(message
)
689 write_string(message
, out
=out
, encoding
=self
.params
.get('encoding'))
691 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
692 """Print message to stdout"""
693 if self
.params
.get('logger'):
694 self
.params
['logger'].debug(message
)
695 elif not quiet
or self
.params
.get('verbose'):
697 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
698 self
._err
_file
if quiet
else self
._screen
_file
)
700 def to_stderr(self
, message
, only_once
=False):
701 """Print message to stderr"""
702 assert isinstance(message
, compat_str
)
703 if self
.params
.get('logger'):
704 self
.params
['logger'].error(message
)
706 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._err
_file
, only_once
=only_once
)
708 def to_console_title(self
, message
):
709 if not self
.params
.get('consoletitle', False):
711 if compat_os_name
== 'nt':
712 if ctypes
.windll
.kernel32
.GetConsoleWindow():
713 # c_wchar_p() might not be necessary if `message` is
714 # already of type unicode()
715 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
716 elif 'TERM' in os
.environ
:
717 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
719 def save_console_title(self
):
720 if not self
.params
.get('consoletitle', False):
722 if self
.params
.get('simulate'):
724 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
725 # Save the title on stack
726 self
._write
_string
('\033[22;0t', self
._screen
_file
)
728 def restore_console_title(self
):
729 if not self
.params
.get('consoletitle', False):
731 if self
.params
.get('simulate'):
733 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
734 # Restore the title from stack
735 self
._write
_string
('\033[23;0t', self
._screen
_file
)
738 self
.save_console_title()
741 def __exit__(self
, *args
):
742 self
.restore_console_title()
744 if self
.params
.get('cookiefile') is not None:
745 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
747 def trouble(self
, message
=None, tb
=None):
748 """Determine action to take when a download problem appears.
750 Depending on if the downloader has been configured to ignore
751 download errors or not, this method may throw an exception or
752 not when errors are found, after printing the message.
754 tb, if given, is additional traceback information.
756 if message
is not None:
757 self
.to_stderr(message
)
758 if self
.params
.get('verbose'):
760 if sys
.exc_info()[0]: # if .trouble has been called from an except block
762 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
763 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
764 tb
+= encode_compat_str(traceback
.format_exc())
766 tb_data
= traceback
.format_list(traceback
.extract_stack())
767 tb
= ''.join(tb_data
)
770 if not self
.params
.get('ignoreerrors', False):
771 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
772 exc_info
= sys
.exc_info()[1].exc_info
774 exc_info
= sys
.exc_info()
775 raise DownloadError(message
, exc_info
)
776 self
._download
_retcode
= 1
778 def to_screen(self
, message
, skip_eol
=False):
779 """Print message to stdout if not in quiet mode"""
781 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
783 def report_warning(self
, message
, only_once
=False):
785 Print the message to stderr, it will be prefixed with 'WARNING:'
786 If stderr is a tty file the 'WARNING:' will be colored
788 if self
.params
.get('logger') is not None:
789 self
.params
['logger'].warning(message
)
791 if self
.params
.get('no_warnings'):
793 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
794 _msg_header
= '\033[0;33mWARNING:\033[0m'
796 _msg_header
= 'WARNING:'
797 warning_message
= '%s %s' % (_msg_header
, message
)
798 self
.to_stderr(warning_message
, only_once
)
800 def report_error(self
, message
, tb
=None):
802 Do the same as trouble, but prefixes the message with 'ERROR:', colored
803 in red if stderr is a tty file.
805 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
806 _msg_header
= '\033[0;31mERROR:\033[0m'
808 _msg_header
= 'ERROR:'
809 error_message
= '%s %s' % (_msg_header
, message
)
810 self
.trouble(error_message
, tb
)
812 def write_debug(self
, message
, only_once
=False):
813 '''Log debug message or Print message to stderr'''
814 if not self
.params
.get('verbose', False):
816 message
= '[debug] %s' % message
817 if self
.params
.get('logger'):
818 self
.params
['logger'].debug(message
)
820 self
.to_stderr(message
, only_once
)
822 def report_file_already_downloaded(self
, file_name
):
823 """Report file has already been fully downloaded."""
825 self
.to_screen('[download] %s has already been downloaded' % file_name
)
826 except UnicodeEncodeError:
827 self
.to_screen('[download] The file has already been downloaded')
829 def report_file_delete(self
, file_name
):
830 """Report that existing file will be deleted."""
832 self
.to_screen('Deleting existing file %s' % file_name
)
833 except UnicodeEncodeError:
834 self
.to_screen('Deleting existing file')
836 def raise_no_formats(self
, has_drm
=False, forced
=False):
837 msg
= 'This video is DRM protected' if has_drm
else 'No video formats found!'
838 expected
= self
.params
.get('ignore_no_formats_error')
839 if forced
or not expected
:
840 raise ExtractorError(msg
, expected
=has_drm
or expected
)
842 self
.report_warning(msg
)
844 def parse_outtmpl(self
):
845 outtmpl_dict
= self
.params
.get('outtmpl', {})
846 if not isinstance(outtmpl_dict
, dict):
847 outtmpl_dict
= {'default': outtmpl_dict}
848 outtmpl_dict
.update({
849 k
: v
for k
, v
in DEFAULT_OUTTMPL
.items()
850 if not outtmpl_dict
.get(k
)})
851 for key
, val
in outtmpl_dict
.items():
852 if isinstance(val
, bytes):
854 'Parameter outtmpl is bytes, but should be a unicode string. '
855 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
858 def get_output_path(self
, dir_type
='', filename
=None):
859 paths
= self
.params
.get('paths', {})
860 assert isinstance(paths
, dict)
862 expand_path(paths
.get('home', '').strip()),
863 expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else '',
866 # Temporary fix for #4787
867 # 'Treat' all problem characters by passing filename through preferredencoding
868 # to workaround encoding issues with subprocess on python2 @ Windows
869 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
870 path
= encodeFilename(path
, True).decode(preferredencoding())
871 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
874 def _outtmpl_expandpath(outtmpl
):
875 # expand_path translates '%%' into '%' and '$$' into '$'
876 # correspondingly that is not what we want since we need to keep
877 # '%%' intact for template dict substitution step. Working around
878 # with boundary-alike separator hack.
879 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
880 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
882 # outtmpl should be expand_path'ed before template dict substitution
883 # because meta fields may contain env variables we don't want to
884 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
885 # title "Hello $PATH", we don't want `$PATH` to be expanded.
886 return expand_path(outtmpl
).replace(sep
, '')
889 def escape_outtmpl(outtmpl
):
890 ''' Escape any remaining strings like %s, %abc% etc. '''
892 STR_FORMAT_RE_TMPL
.format('', '(?![%(\0])'),
893 lambda mobj
: ('' if mobj
.group('has_key') else '%') + mobj
.group(0),
897 def validate_outtmpl(cls
, outtmpl
):
898 ''' @return None or Exception object '''
900 STR_FORMAT_RE_TMPL
.format('[^)]*', '[ljq]'),
901 lambda mobj
: f
'{mobj.group(0)[:-1]}s',
902 cls
._outtmpl
_expandpath
(outtmpl
))
904 cls
.escape_outtmpl(outtmpl
) % collections
.defaultdict(int)
906 except ValueError as err
:
909 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=None):
910 """ Make the template and info_dict suitable for substitution : ydl.outtmpl_escape(outtmpl) % info_dict """
911 info_dict
.setdefault('epoch', int(time
.time())) # keep epoch consistent once set
913 info_dict
= dict(info_dict
) # Do not sanitize so as not to consume LazyList
914 for key
in ('__original_infodict', '__postprocessors'):
915 info_dict
.pop(key
, None)
916 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
917 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
918 if info_dict
.get('duration', None) is not None
920 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
921 if info_dict
.get('resolution') is None:
922 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
924 # For fields playlist_index and autonumber convert all occurrences
925 # of %(field)s to %(field)0Nd for backward compatibility
926 field_size_compat_map
= {
927 'playlist_index': len(str(info_dict
.get('_last_playlist_index') or '')),
928 'autonumber': self
.params
.get('autonumber_size') or 5,
932 EXTERNAL_FORMAT_RE
= re
.compile(STR_FORMAT_RE_TMPL
.format('[^)]*', f
'[{STR_FORMAT_TYPES}ljq]'))
937 # Field is of the form key1.key2...
938 # where keys (except first) can be string, int or slice
939 FIELD_RE
= r
'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num
=r
'(?:-?\d+)')
940 MATH_FIELD_RE
= r
'''{field}|{num}'''.format(field
=FIELD_RE
, num
=r
'-?\d+(?:.\d+)?')
941 MATH_OPERATORS_RE
= r
'(?:%s)' % '|'.join(map(re
.escape
, MATH_FUNCTIONS
.keys()))
942 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
945 (?P<maths>(?:{math_op}{math_field})*)
946 (?:>(?P<strf_format>.+?))?
947 (?:\|(?P<default>.*?))?
948 $'''.format(field
=FIELD_RE
, math_op
=MATH_OPERATORS_RE
, math_field
=MATH_FIELD_RE
))
950 def _traverse_infodict(k
):
954 return traverse_obj(info_dict
, k
, is_user_input
=True, traverse_string
=True)
956 def get_value(mdict
):
958 value
= _traverse_infodict(mdict
['fields'])
961 value
= float_or_none(value
)
962 if value
is not None:
965 offset_key
= mdict
['maths']
967 value
= float_or_none(value
)
971 MATH_FIELD_RE
if operator
else MATH_OPERATORS_RE
,
973 offset_key
= offset_key
[len(item
):]
975 operator
= MATH_FUNCTIONS
[item
]
977 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
978 offset
= float_or_none(item
)
980 offset
= float_or_none(_traverse_infodict(item
))
982 value
= operator(value
, multiplier
* offset
)
983 except (TypeError, ZeroDivisionError):
986 # Datetime formatting
987 if mdict
['strf_format']:
988 value
= strftime_or_none(value
, mdict
['strf_format'])
992 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
994 def _dumpjson_default(obj
):
995 if isinstance(obj
, (set, LazyList
)):
997 raise TypeError(f
'Object of type {type(obj).__name__} is not JSON serializable')
999 def create_key(outer_mobj
):
1000 if not outer_mobj
.group('has_key'):
1001 return f
'%{outer_mobj.group(0)}'
1002 key
= outer_mobj
.group('key')
1003 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
1005 value
, default
, mobj
= None, na
, {'fields': ''}
1007 mobj
= mobj
.groupdict()
1008 default
= mobj
['default'] if mobj
['default'] is not None else na
1009 value
= get_value(mobj
)
1011 fmt
= outer_mobj
.group('format')
1012 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
1013 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
1015 value
= default
if value
is None else value
1017 str_fmt
= f
'{fmt[:-1]}s'
1019 value
, fmt
= ', '.join(variadic(value
)), str_fmt
1020 elif fmt
[-1] == 'j':
1021 value
, fmt
= json
.dumps(value
, default
=_dumpjson_default
), str_fmt
1022 elif fmt
[-1] == 'q':
1023 value
, fmt
= compat_shlex_quote(str(value
)), str_fmt
1024 elif fmt
[-1] == 'c':
1027 value
, fmt
= default
, 's'
1030 elif fmt
[-1] not in 'rs': # numeric
1031 value
= float_or_none(value
)
1033 value
, fmt
= default
, 's'
1037 # If value is an object, sanitize might convert it to a string
1038 # So we convert it to repr first
1039 value
, fmt
= repr(value
), str_fmt
1040 if fmt
[-1] in 'csr':
1041 value
= sanitize(mobj
['fields'].split('.')[-1], value
)
1043 key
= '%s\0%s' % (key
.replace('%', '%\0'), outer_mobj
.group('format'))
1044 TMPL_DICT
[key
] = value
1045 return '{prefix}%({key}){fmt}'.format(key
=key
, fmt
=fmt
, prefix
=outer_mobj
.group('prefix'))
1047 return EXTERNAL_FORMAT_RE
.sub(create_key
, outtmpl
), TMPL_DICT
1049 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
1051 sanitize
= lambda k
, v
: sanitize_filename(
1053 restricted
=self
.params
.get('restrictfilenames'),
1054 is_id
=(k
== 'id' or k
.endswith('_id')))
1055 outtmpl
= self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default'])
1056 outtmpl
, template_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, sanitize
)
1057 outtmpl
= self
.escape_outtmpl(self
._outtmpl
_expandpath
(outtmpl
))
1058 filename
= outtmpl
% template_dict
1060 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
1061 if force_ext
is not None:
1062 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
1064 # https://github.com/blackjack4494/youtube-dlc/issues/85
1065 trim_file_name
= self
.params
.get('trim_file_name', False)
1067 fn_groups
= filename
.rsplit('.')
1070 if len(fn_groups
) > 2:
1071 sub_ext
= fn_groups
[-2]
1072 filename
= '.'.join(filter(None, [fn_groups
[0][:trim_file_name
], sub_ext
, ext
]))
1075 except ValueError as err
:
1076 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
1079 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
1080 """Generate the output filename."""
1082 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
1085 if not self
.params
.get('paths'):
1087 elif filename
== '-':
1088 self
.report_warning('--paths is ignored when an outputting to stdout', only_once
=True)
1089 elif os
.path
.isabs(filename
):
1090 self
.report_warning('--paths is ignored since an absolute path is given in output template', only_once
=True)
1091 if filename
== '-' or not filename
:
1094 return self
.get_output_path(dir_type
, filename
)
1096 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1097 """ Returns None if the file should be downloaded """
1099 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1102 if 'title' in info_dict
:
1103 # This can happen when we're just evaluating the playlist
1104 title
= info_dict
['title']
1105 matchtitle
= self
.params
.get('matchtitle', False)
1107 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1108 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1109 rejecttitle
= self
.params
.get('rejecttitle', False)
1111 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1112 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1113 date
= info_dict
.get('upload_date')
1114 if date
is not None:
1115 dateRange
= self
.params
.get('daterange', DateRange())
1116 if date
not in dateRange
:
1117 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1118 view_count
= info_dict
.get('view_count')
1119 if view_count
is not None:
1120 min_views
= self
.params
.get('min_views')
1121 if min_views
is not None and view_count
< min_views
:
1122 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1123 max_views
= self
.params
.get('max_views')
1124 if max_views
is not None and view_count
> max_views
:
1125 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1126 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1127 return 'Skipping "%s" because it is age restricted' % video_title
1129 match_filter
= self
.params
.get('match_filter')
1130 if match_filter
is not None:
1132 ret
= match_filter(info_dict
, incomplete
=incomplete
)
1134 # For backward compatibility
1135 ret
= None if incomplete
else match_filter(info_dict
)
1140 if self
.in_download_archive(info_dict
):
1141 reason
= '%s has already been recorded in the archive' % video_title
1142 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1144 reason
= check_filter()
1145 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1146 if reason
is not None:
1148 self
.to_screen('[download] ' + reason
)
1149 if self
.params
.get(break_opt
, False):
1154 def add_extra_info(info_dict
, extra_info
):
1155 '''Set the keys from extra_info in info dict if they are missing'''
1156 for key
, value
in extra_info
.items():
1157 info_dict
.setdefault(key
, value
)
1159 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
={},
1160 process
=True, force_generic_extractor
=False):
1162 Return a list with a dictionary for each video extracted.
1165 url -- URL to extract
1168 download -- whether to download videos during extraction
1169 ie_key -- extractor key hint
1170 extra_info -- dictionary containing the extra values to add to each result
1171 process -- whether to resolve all unresolved references (URLs, playlist items),
1172 must be True for download to work.
1173 force_generic_extractor -- force using the generic extractor
1176 if not ie_key
and force_generic_extractor
:
1180 ies
= [self
.get_info_extractor(ie_key
)]
1185 if not ie
.suitable(url
):
1188 ie_key
= ie
.ie_key()
1189 ie
= self
.get_info_extractor(ie_key
)
1190 if not ie
.working():
1191 self
.report_warning('The program functionality for this site has been marked as broken, '
1192 'and will probably not work.')
1195 temp_id
= str_or_none(
1196 ie
.extract_id(url
) if callable(getattr(ie
, 'extract_id', None))
1197 else ie
._match
_id
(url
))
1198 except (AssertionError, IndexError, AttributeError):
1200 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1201 self
.to_screen("[%s] %s: has already been recorded in archive" % (
1204 return self
.__extract
_info
(url
, ie
, download
, extra_info
, process
)
1206 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1208 def __handle_extraction_exceptions(func
, handle_all_errors
=True):
1209 def wrapper(self
, *args
, **kwargs
):
1211 return func(self
, *args
, **kwargs
)
1212 except GeoRestrictedError
as e
:
1215 msg
+= '\nThis video is available in %s.' % ', '.join(
1216 map(ISO3166Utils
.short2full
, e
.countries
))
1217 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1218 self
.report_error(msg
)
1219 except ExtractorError
as e
: # An error we somewhat expected
1220 self
.report_error(compat_str(e
), e
.format_traceback())
1221 except ThrottledDownload
:
1222 self
.to_stderr('\r')
1223 self
.report_warning('The download speed is below throttle limit. Re-extracting data')
1224 return wrapper(self
, *args
, **kwargs
)
1225 except (MaxDownloadsReached
, ExistingVideoReached
, RejectedVideoReached
):
1227 except Exception as e
:
1228 if handle_all_errors
and self
.params
.get('ignoreerrors', False):
1229 self
.report_error(error_to_compat_str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1234 @__handle_extraction_exceptions
1235 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1236 ie_result
= ie
.extract(url
)
1237 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1239 if isinstance(ie_result
, list):
1240 # Backwards compatibility: old IE result format
1242 '_type': 'compat_list',
1243 'entries': ie_result
,
1245 if extra_info
.get('original_url'):
1246 ie_result
.setdefault('original_url', extra_info
['original_url'])
1247 self
.add_default_extra_info(ie_result
, ie
, url
)
1249 return self
.process_ie_result(ie_result
, download
, extra_info
)
1253 def add_default_extra_info(self
, ie_result
, ie
, url
):
1255 self
.add_extra_info(ie_result
, {
1257 'original_url': url
,
1258 'webpage_url_basename': url_basename(url
),
1261 self
.add_extra_info(ie_result
, {
1262 'extractor': ie
.IE_NAME
,
1263 'extractor_key': ie
.ie_key(),
1266 def process_ie_result(self
, ie_result
, download
=True, extra_info
=None):
1268 Take the result of the ie(may be modified) and resolve all unresolved
1269 references (URLs, playlist items).
1271 It will also download the videos if 'download'.
1272 Returns the resolved ie_result.
1274 if extra_info
is None:
1276 result_type
= ie_result
.get('_type', 'video')
1278 if result_type
in ('url', 'url_transparent'):
1279 ie_result
['url'] = sanitize_url(ie_result
['url'])
1280 if ie_result
.get('original_url'):
1281 extra_info
.setdefault('original_url', ie_result
['original_url'])
1283 extract_flat
= self
.params
.get('extract_flat', False)
1284 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1285 or extract_flat
is True):
1286 info_copy
= ie_result
.copy()
1287 self
.add_extra_info(info_copy
, extra_info
)
1288 ie
= try_get(ie_result
.get('ie_key'), self
.get_info_extractor
)
1289 self
.add_default_extra_info(info_copy
, ie
, ie_result
['url'])
1290 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1293 if result_type
== 'video':
1294 self
.add_extra_info(ie_result
, extra_info
)
1295 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1296 additional_urls
= (ie_result
or {}).get('additional_urls')
1298 # TODO: Improve MetadataParserPP to allow setting a list
1299 if isinstance(additional_urls
, compat_str
):
1300 additional_urls
= [additional_urls
]
1302 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1303 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1304 ie_result
['additional_entries'] = [
1306 url
, download
, extra_info
,
1307 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1308 for url
in additional_urls
1311 elif result_type
== 'url':
1312 # We have to add extra_info to the results because it may be
1313 # contained in a playlist
1314 return self
.extract_info(
1315 ie_result
['url'], download
,
1316 ie_key
=ie_result
.get('ie_key'),
1317 extra_info
=extra_info
)
1318 elif result_type
== 'url_transparent':
1319 # Use the information from the embedding page
1320 info
= self
.extract_info(
1321 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1322 extra_info
=extra_info
, download
=False, process
=False)
1324 # extract_info may return None when ignoreerrors is enabled and
1325 # extraction failed with an error, don't crash and return early
1330 force_properties
= dict(
1331 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1332 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1333 if f
in force_properties
:
1334 del force_properties
[f
]
1335 new_result
= info
.copy()
1336 new_result
.update(force_properties
)
1338 # Extracted info may not be a video result (i.e.
1339 # info.get('_type', 'video') != video) but rather an url or
1340 # url_transparent. In such cases outer metadata (from ie_result)
1341 # should be propagated to inner one (info). For this to happen
1342 # _type of info should be overridden with url_transparent. This
1343 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1344 if new_result
.get('_type') == 'url':
1345 new_result
['_type'] = 'url_transparent'
1347 return self
.process_ie_result(
1348 new_result
, download
=download
, extra_info
=extra_info
)
1349 elif result_type
in ('playlist', 'multi_video'):
1350 # Protect from infinite recursion due to recursively nested playlists
1351 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1352 webpage_url
= ie_result
['webpage_url']
1353 if webpage_url
in self
._playlist
_urls
:
1355 '[download] Skipping already downloaded playlist: %s'
1356 % ie_result
.get('title') or ie_result
.get('id'))
1359 self
._playlist
_level
+= 1
1360 self
._playlist
_urls
.add(webpage_url
)
1361 self
._sanitize
_thumbnails
(ie_result
)
1363 return self
.__process
_playlist
(ie_result
, download
)
1365 self
._playlist
_level
-= 1
1366 if not self
._playlist
_level
:
1367 self
._playlist
_urls
.clear()
1368 elif result_type
== 'compat_list':
1369 self
.report_warning(
1370 'Extractor %s returned a compat_list result. '
1371 'It needs to be updated.' % ie_result
.get('extractor'))
1374 self
.add_extra_info(r
, {
1375 'extractor': ie_result
['extractor'],
1376 'webpage_url': ie_result
['webpage_url'],
1377 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1378 'extractor_key': ie_result
['extractor_key'],
1381 ie_result
['entries'] = [
1382 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1383 for r
in ie_result
['entries']
1387 raise Exception('Invalid result type: %s' % result_type
)
1389 def _ensure_dir_exists(self
, path
):
1390 return make_dir(path
, self
.report_error
)
1392 def __process_playlist(self
, ie_result
, download
):
1393 # We process each entry in the playlist
1394 playlist
= ie_result
.get('title') or ie_result
.get('id')
1395 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1397 if 'entries' not in ie_result
:
1398 raise EntryNotInPlaylist()
1399 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1400 if incomplete_entries
:
1401 def fill_missing_entries(entries
, indexes
):
1402 ret
= [None] * max(*indexes
)
1403 for i
, entry
in zip(indexes
, entries
):
1406 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1408 playlist_results
= []
1410 playliststart
= self
.params
.get('playliststart', 1)
1411 playlistend
= self
.params
.get('playlistend')
1412 # For backwards compatibility, interpret -1 as whole list
1413 if playlistend
== -1:
1416 playlistitems_str
= self
.params
.get('playlist_items')
1417 playlistitems
= None
1418 if playlistitems_str
is not None:
1419 def iter_playlistitems(format
):
1420 for string_segment
in format
.split(','):
1421 if '-' in string_segment
:
1422 start
, end
= string_segment
.split('-')
1423 for item
in range(int(start
), int(end
) + 1):
1426 yield int(string_segment
)
1427 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1429 ie_entries
= ie_result
['entries']
1431 'Downloading %d videos' if not isinstance(ie_entries
, list)
1432 else 'Collected %d videos; downloading %%d of them' % len(ie_entries
))
1433 if not isinstance(ie_entries
, (list, PagedList
)):
1434 ie_entries
= LazyList(ie_entries
)
1437 return YoutubeDL
.__handle
_extraction
_exceptions
(
1438 lambda self
, i
: ie_entries
[i
- 1],
1443 for i
in playlistitems
or itertools
.count(playliststart
):
1444 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1448 entry
= get_entry(i
)
1450 raise EntryNotInPlaylist()
1451 except (IndexError, EntryNotInPlaylist
):
1452 if incomplete_entries
:
1453 raise EntryNotInPlaylist()
1454 elif not playlistitems
:
1456 entries
.append(entry
)
1458 if entry
is not None:
1459 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1460 except (ExistingVideoReached
, RejectedVideoReached
):
1462 ie_result
['entries'] = entries
1464 # Save playlist_index before re-ordering
1466 ((playlistitems
[i
- 1] if playlistitems
else i
+ playliststart
- 1), entry
)
1467 for i
, entry
in enumerate(entries
, 1)
1468 if entry
is not None]
1469 n_entries
= len(entries
)
1471 if not playlistitems
and (playliststart
or playlistend
):
1472 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1473 ie_result
['requested_entries'] = playlistitems
1475 if self
.params
.get('allow_playlist_files', True):
1477 'playlist': playlist
,
1478 'playlist_id': ie_result
.get('id'),
1479 'playlist_title': ie_result
.get('title'),
1480 'playlist_uploader': ie_result
.get('uploader'),
1481 'playlist_uploader_id': ie_result
.get('uploader_id'),
1482 'playlist_index': 0,
1484 ie_copy
.update(dict(ie_result
))
1486 if self
.params
.get('writeinfojson', False):
1487 infofn
= self
.prepare_filename(ie_copy
, 'pl_infojson')
1488 if not self
._ensure
_dir
_exists
(encodeFilename(infofn
)):
1490 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(infofn
)):
1491 self
.to_screen('[info] Playlist metadata is already present')
1493 self
.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn
)
1495 write_json_file(self
.sanitize_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
1496 except (OSError, IOError):
1497 self
.report_error('Cannot write playlist metadata to JSON file ' + infofn
)
1499 # TODO: This should be passed to ThumbnailsConvertor if necessary
1500 self
._write
_thumbnails
(ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1502 if self
.params
.get('writedescription', False):
1503 descfn
= self
.prepare_filename(ie_copy
, 'pl_description')
1504 if not self
._ensure
_dir
_exists
(encodeFilename(descfn
)):
1506 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(descfn
)):
1507 self
.to_screen('[info] Playlist description is already present')
1508 elif ie_result
.get('description') is None:
1509 self
.report_warning('There\'s no playlist description to write.')
1512 self
.to_screen('[info] Writing playlist description to: ' + descfn
)
1513 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
1514 descfile
.write(ie_result
['description'])
1515 except (OSError, IOError):
1516 self
.report_error('Cannot write playlist description file ' + descfn
)
1519 if self
.params
.get('playlistreverse', False):
1520 entries
= entries
[::-1]
1521 if self
.params
.get('playlistrandom', False):
1522 random
.shuffle(entries
)
1524 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1526 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1528 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1529 for i
, entry_tuple
in enumerate(entries
, 1):
1530 playlist_index
, entry
= entry_tuple
1531 if 'playlist-index' in self
.params
.get('compat_options', []):
1532 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
1533 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1534 # This __x_forwarded_for_ip thing is a bit ugly but requires
1537 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1539 'n_entries': n_entries
,
1540 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1541 'playlist_index': playlist_index
,
1542 'playlist_autonumber': i
,
1543 'playlist': playlist
,
1544 'playlist_id': ie_result
.get('id'),
1545 'playlist_title': ie_result
.get('title'),
1546 'playlist_uploader': ie_result
.get('uploader'),
1547 'playlist_uploader_id': ie_result
.get('uploader_id'),
1548 'extractor': ie_result
['extractor'],
1549 'webpage_url': ie_result
['webpage_url'],
1550 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1551 'extractor_key': ie_result
['extractor_key'],
1554 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1557 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1558 if not entry_result
:
1560 if failures
>= max_failures
:
1562 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1564 # TODO: skip failed (empty) entries?
1565 playlist_results
.append(entry_result
)
1566 ie_result
['entries'] = playlist_results
1567 self
.to_screen('[download] Finished downloading playlist: %s' % playlist
)
1570 @__handle_extraction_exceptions
1571 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1572 return self
.process_ie_result(
1573 entry
, download
=download
, extra_info
=extra_info
)
1575 def _build_format_filter(self
, filter_spec
):
1576 " Returns a function to filter the formats according to the filter_spec "
1586 operator_rex
= re
.compile(r
'''(?x)\s*
1587 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1588 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1589 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1590 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1591 m
= operator_rex
.fullmatch(filter_spec
)
1594 comparison_value
= int(m
.group('value'))
1596 comparison_value
= parse_filesize(m
.group('value'))
1597 if comparison_value
is None:
1598 comparison_value
= parse_filesize(m
.group('value') + 'B')
1599 if comparison_value
is None:
1601 'Invalid value %r in format specification %r' % (
1602 m
.group('value'), filter_spec
))
1603 op
= OPERATORS
[m
.group('op')]
1608 '^=': lambda attr
, value
: attr
.startswith(value
),
1609 '$=': lambda attr
, value
: attr
.endswith(value
),
1610 '*=': lambda attr
, value
: value
in attr
,
1612 str_operator_rex
= re
.compile(r
'''(?x)\s*
1613 (?P<key>[a-zA-Z0-9._-]+)\s*
1614 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1615 (?P<value>[a-zA-Z0-9._-]+)\s*
1616 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1617 m
= str_operator_rex
.fullmatch(filter_spec
)
1619 comparison_value
= m
.group('value')
1620 str_op
= STR_OPERATORS
[m
.group('op')]
1621 if m
.group('negation'):
1622 op
= lambda attr
, value
: not str_op(attr
, value
)
1627 raise SyntaxError('Invalid filter specification %r' % filter_spec
)
1630 actual_value
= f
.get(m
.group('key'))
1631 if actual_value
is None:
1632 return m
.group('none_inclusive')
1633 return op(actual_value
, comparison_value
)
1636 def _default_format_spec(self
, info_dict
, download
=True):
1639 merger
= FFmpegMergerPP(self
)
1640 return merger
.available
and merger
.can_merge()
1643 not self
.params
.get('simulate')
1647 or info_dict
.get('is_live', False)
1648 or self
.outtmpl_dict
['default'] == '-'))
1651 or self
.params
.get('allow_multiple_audio_streams', False)
1652 or 'format-spec' in self
.params
.get('compat_opts', []))
1655 'best/bestvideo+bestaudio' if prefer_best
1656 else 'bestvideo*+bestaudio/best' if not compat
1657 else 'bestvideo+bestaudio/best')
1659 def build_format_selector(self
, format_spec
):
1660 def syntax_error(note
, start
):
1662 'Invalid format specification: '
1663 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1664 return SyntaxError(message
)
1666 PICKFIRST
= 'PICKFIRST'
1670 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1672 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1673 'video': self
.params
.get('allow_multiple_video_streams', False)}
1675 check_formats
= self
.params
.get('check_formats')
1677 def _parse_filter(tokens
):
1679 for type, string
, start
, _
, _
in tokens
:
1680 if type == tokenize
.OP
and string
== ']':
1681 return ''.join(filter_parts
)
1683 filter_parts
.append(string
)
1685 def _remove_unused_ops(tokens
):
1686 # Remove operators that we don't use and join them with the surrounding strings
1687 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1688 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1689 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1690 for type, string
, start
, end
, line
in tokens
:
1691 if type == tokenize
.OP
and string
== '[':
1693 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1695 yield type, string
, start
, end
, line
1696 # everything inside brackets will be handled by _parse_filter
1697 for type, string
, start
, end
, line
in tokens
:
1698 yield type, string
, start
, end
, line
1699 if type == tokenize
.OP
and string
== ']':
1701 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1703 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1705 yield type, string
, start
, end
, line
1706 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1708 last_string
= string
1712 last_string
+= string
1714 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1716 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1718 current_selector
= None
1719 for type, string
, start
, _
, _
in tokens
:
1720 # ENCODING is only defined in python 3.x
1721 if type == getattr(tokenize
, 'ENCODING', None):
1723 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1724 current_selector
= FormatSelector(SINGLE
, string
, [])
1725 elif type == tokenize
.OP
:
1727 if not inside_group
:
1728 # ')' will be handled by the parentheses group
1729 tokens
.restore_last_token()
1731 elif inside_merge
and string
in ['/', ',']:
1732 tokens
.restore_last_token()
1734 elif inside_choice
and string
== ',':
1735 tokens
.restore_last_token()
1738 if not current_selector
:
1739 raise syntax_error('"," must follow a format selector', start
)
1740 selectors
.append(current_selector
)
1741 current_selector
= None
1743 if not current_selector
:
1744 raise syntax_error('"/" must follow a format selector', start
)
1745 first_choice
= current_selector
1746 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1747 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1749 if not current_selector
:
1750 current_selector
= FormatSelector(SINGLE
, 'best', [])
1751 format_filter
= _parse_filter(tokens
)
1752 current_selector
.filters
.append(format_filter
)
1754 if current_selector
:
1755 raise syntax_error('Unexpected "("', start
)
1756 group
= _parse_format_selection(tokens
, inside_group
=True)
1757 current_selector
= FormatSelector(GROUP
, group
, [])
1759 if not current_selector
:
1760 raise syntax_error('Unexpected "+"', start
)
1761 selector_1
= current_selector
1762 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1764 raise syntax_error('Expected a selector', start
)
1765 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
1767 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
1768 elif type == tokenize
.ENDMARKER
:
1770 if current_selector
:
1771 selectors
.append(current_selector
)
1774 def _merge(formats_pair
):
1775 format_1
, format_2
= formats_pair
1778 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
1779 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
1781 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
1782 get_no_more
= {'video': False, 'audio': False}
1783 for (i
, fmt_info
) in enumerate(formats_info
):
1784 if fmt_info
.get('acodec') == fmt_info
.get('vcodec') == 'none':
1787 for aud_vid
in ['audio', 'video']:
1788 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
1789 if get_no_more
[aud_vid
]:
1792 get_no_more
[aud_vid
] = True
1794 if len(formats_info
) == 1:
1795 return formats_info
[0]
1797 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
1798 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
1800 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
1801 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
1803 output_ext
= self
.params
.get('merge_output_format')
1806 output_ext
= the_only_video
['ext']
1807 elif the_only_audio
and not video_fmts
:
1808 output_ext
= the_only_audio
['ext']
1813 'requested_formats': formats_info
,
1814 'format': '+'.join(fmt_info
.get('format') for fmt_info
in formats_info
),
1815 'format_id': '+'.join(fmt_info
.get('format_id') for fmt_info
in formats_info
),
1821 'width': the_only_video
.get('width'),
1822 'height': the_only_video
.get('height'),
1823 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
1824 'fps': the_only_video
.get('fps'),
1825 'vcodec': the_only_video
.get('vcodec'),
1826 'vbr': the_only_video
.get('vbr'),
1827 'stretched_ratio': the_only_video
.get('stretched_ratio'),
1832 'acodec': the_only_audio
.get('acodec'),
1833 'abr': the_only_audio
.get('abr'),
1838 def _check_formats(formats
):
1839 if not check_formats
:
1843 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1844 temp_file
= tempfile
.NamedTemporaryFile(
1845 suffix
='.tmp', delete
=False,
1846 dir=self
.get_output_path('temp') or None)
1849 success
, _
= self
.dl(temp_file
.name
, f
, test
=True)
1850 except (DownloadError
, IOError, OSError, ValueError) + network_exceptions
:
1853 if os
.path
.exists(temp_file
.name
):
1855 os
.remove(temp_file
.name
)
1857 self
.report_warning('Unable to delete temporary file "%s"' % temp_file
.name
)
1861 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1863 def _build_selector_function(selector
):
1864 if isinstance(selector
, list): # ,
1865 fs
= [_build_selector_function(s
) for s
in selector
]
1867 def selector_function(ctx
):
1870 return selector_function
1872 elif selector
.type == GROUP
: # ()
1873 selector_function
= _build_selector_function(selector
.selector
)
1875 elif selector
.type == PICKFIRST
: # /
1876 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
1878 def selector_function(ctx
):
1880 picked_formats
= list(f(ctx
))
1882 return picked_formats
1885 elif selector
.type == MERGE
: # +
1886 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
1888 def selector_function(ctx
):
1889 for pair
in itertools
.product(
1890 selector_1(copy
.deepcopy(ctx
)), selector_2(copy
.deepcopy(ctx
))):
1893 elif selector
.type == SINGLE
: # atom
1894 format_spec
= selector
.selector
or 'best'
1896 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
1897 if format_spec
== 'all':
1898 def selector_function(ctx
):
1899 yield from _check_formats(ctx
['formats'])
1900 elif format_spec
== 'mergeall':
1901 def selector_function(ctx
):
1902 formats
= list(_check_formats(ctx
['formats']))
1905 merged_format
= formats
[-1]
1906 for f
in formats
[-2::-1]:
1907 merged_format
= _merge((merged_format
, f
))
1911 format_fallback
, format_reverse
, format_idx
= False, True, 1
1913 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1915 if mobj
is not None:
1916 format_idx
= int_or_none(mobj
.group('n'), default
=1)
1917 format_reverse
= mobj
.group('bw')[0] == 'b'
1918 format_type
= (mobj
.group('type') or [None])[0]
1919 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
1920 format_modified
= mobj
.group('mod') is not None
1922 format_fallback
= not format_type
and not format_modified
# for b, w
1924 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
1925 if format_type
and format_modified
# bv*, ba*, wv*, wa*
1926 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
1927 if format_type
# bv, ba, wv, wa
1928 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
1929 if not format_modified
# b, w
1930 else lambda f
: True) # b*, w*
1931 filter_f
= lambda f
: _filter_f(f
) and (
1932 f
.get('vcodec') != 'none' or f
.get('acodec') != 'none')
1934 filter_f
= ((lambda f
: f
.get('ext') == format_spec
)
1935 if format_spec
in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1936 else (lambda f
: f
.get('format_id') == format_spec
)) # id
1938 def selector_function(ctx
):
1939 formats
= list(ctx
['formats'])
1940 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
1941 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
1942 # for extractors with incomplete formats (audio only (soundcloud)
1943 # or video only (imgur)) best/worst will fallback to
1944 # best/worst {video,audio}-only format
1946 matches
= LazyList(_check_formats(matches
[::-1 if format_reverse
else 1]))
1948 yield matches
[format_idx
- 1]
1952 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
1954 def final_selector(ctx
):
1955 ctx_copy
= copy
.deepcopy(ctx
)
1956 for _filter
in filters
:
1957 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
1958 return selector_function(ctx_copy
)
1959 return final_selector
1961 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
1963 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
1964 except tokenize
.TokenError
:
1965 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
1967 class TokenIterator(object):
1968 def __init__(self
, tokens
):
1969 self
.tokens
= tokens
1976 if self
.counter
>= len(self
.tokens
):
1977 raise StopIteration()
1978 value
= self
.tokens
[self
.counter
]
1984 def restore_last_token(self
):
1987 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
1988 return _build_selector_function(parsed_selector
)
1990 def _calc_headers(self
, info_dict
):
1991 res
= std_headers
.copy()
1993 add_headers
= info_dict
.get('http_headers')
1995 res
.update(add_headers
)
1997 cookies
= self
._calc
_cookies
(info_dict
)
1999 res
['Cookie'] = cookies
2001 if 'X-Forwarded-For' not in res
:
2002 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
2003 if x_forwarded_for_ip
:
2004 res
['X-Forwarded-For'] = x_forwarded_for_ip
2008 def _calc_cookies(self
, info_dict
):
2009 pr
= sanitized_Request(info_dict
['url'])
2010 self
.cookiejar
.add_cookie_header(pr
)
2011 return pr
.get_header('Cookie')
2013 def _sanitize_thumbnails(self
, info_dict
):
2014 thumbnails
= info_dict
.get('thumbnails')
2015 if thumbnails
is None:
2016 thumbnail
= info_dict
.get('thumbnail')
2018 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
2020 thumbnails
.sort(key
=lambda t
: (
2021 t
.get('preference') if t
.get('preference') is not None else -1,
2022 t
.get('width') if t
.get('width') is not None else -1,
2023 t
.get('height') if t
.get('height') is not None else -1,
2024 t
.get('id') if t
.get('id') is not None else '',
2027 def thumbnail_tester():
2028 if self
.params
.get('check_formats'):
2030 to_screen
= lambda msg
: self
.to_screen(f
'[info] {msg}')
2033 to_screen
= self
.write_debug
2035 def test_thumbnail(t
):
2036 if not test_all
and not t
.get('_test_url'):
2038 to_screen('Testing thumbnail %s' % t
['id'])
2040 self
.urlopen(HEADRequest(t
['url']))
2041 except network_exceptions
as err
:
2042 to_screen('Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % (
2043 t
['id'], t
['url'], error_to_compat_str(err
)))
2047 return test_thumbnail
2049 for i
, t
in enumerate(thumbnails
):
2050 if t
.get('id') is None:
2052 if t
.get('width') and t
.get('height'):
2053 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
2054 t
['url'] = sanitize_url(t
['url'])
2056 if self
.params
.get('check_formats') is not False:
2057 info_dict
['thumbnails'] = LazyList(filter(thumbnail_tester(), thumbnails
[::-1])).reverse()
2059 info_dict
['thumbnails'] = thumbnails
2061 def process_video_result(self
, info_dict
, download
=True):
2062 assert info_dict
.get('_type', 'video') == 'video'
2064 if 'id' not in info_dict
:
2065 raise ExtractorError('Missing "id" field in extractor result')
2066 if 'title' not in info_dict
:
2067 raise ExtractorError('Missing "title" field in extractor result')
2069 def report_force_conversion(field
, field_not
, conversion
):
2070 self
.report_warning(
2071 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2072 % (field
, field_not
, conversion
))
2074 def sanitize_string_field(info
, string_field
):
2075 field
= info
.get(string_field
)
2076 if field
is None or isinstance(field
, compat_str
):
2078 report_force_conversion(string_field
, 'a string', 'string')
2079 info
[string_field
] = compat_str(field
)
2081 def sanitize_numeric_fields(info
):
2082 for numeric_field
in self
._NUMERIC
_FIELDS
:
2083 field
= info
.get(numeric_field
)
2084 if field
is None or isinstance(field
, compat_numeric_types
):
2086 report_force_conversion(numeric_field
, 'numeric', 'int')
2087 info
[numeric_field
] = int_or_none(field
)
2089 sanitize_string_field(info_dict
, 'id')
2090 sanitize_numeric_fields(info_dict
)
2092 if 'playlist' not in info_dict
:
2093 # It isn't part of a playlist
2094 info_dict
['playlist'] = None
2095 info_dict
['playlist_index'] = None
2097 self
._sanitize
_thumbnails
(info_dict
)
2099 thumbnail
= info_dict
.get('thumbnail')
2100 thumbnails
= info_dict
.get('thumbnails')
2102 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
2104 info_dict
['thumbnail'] = thumbnails
[-1]['url']
2106 if info_dict
.get('display_id') is None and 'id' in info_dict
:
2107 info_dict
['display_id'] = info_dict
['id']
2109 for ts_key
, date_key
in (
2110 ('timestamp', 'upload_date'),
2111 ('release_timestamp', 'release_date'),
2113 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
2114 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2115 # see http://bugs.python.org/issue1646728)
2117 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
2118 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
2119 except (ValueError, OverflowError, OSError):
2122 live_keys
= ('is_live', 'was_live')
2123 live_status
= info_dict
.get('live_status')
2124 if live_status
is None:
2125 for key
in live_keys
:
2126 if info_dict
.get(key
) is False:
2128 if info_dict
.get(key
):
2131 if all(info_dict
.get(key
) is False for key
in live_keys
):
2132 live_status
= 'not_live'
2134 info_dict
['live_status'] = live_status
2135 for key
in live_keys
:
2136 if info_dict
.get(key
) is None:
2137 info_dict
[key
] = (live_status
== key
)
2139 # Auto generate title fields corresponding to the *_number fields when missing
2140 # in order to always have clean titles. This is very common for TV series.
2141 for field
in ('chapter', 'season', 'episode'):
2142 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2143 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2145 for cc_kind
in ('subtitles', 'automatic_captions'):
2146 cc
= info_dict
.get(cc_kind
)
2148 for _
, subtitle
in cc
.items():
2149 for subtitle_format
in subtitle
:
2150 if subtitle_format
.get('url'):
2151 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2152 if subtitle_format
.get('ext') is None:
2153 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2155 automatic_captions
= info_dict
.get('automatic_captions')
2156 subtitles
= info_dict
.get('subtitles')
2158 info_dict
['requested_subtitles'] = self
.process_subtitles(
2159 info_dict
['id'], subtitles
, automatic_captions
)
2161 # We now pick which formats have to be downloaded
2162 if info_dict
.get('formats') is None:
2163 # There's only one format available
2164 formats
= [info_dict
]
2166 formats
= info_dict
['formats']
2168 if not self
.params
.get('allow_unplayable_formats'):
2169 formats
= [f
for f
in formats
if not f
.get('has_drm')]
2170 info_dict
['__has_drm'] = len(info_dict
.get('formats') or ['']) > len(formats
)
2173 self
.raise_no_formats(info_dict
.get('__has_drm'))
2175 def is_wellformed(f
):
2178 self
.report_warning(
2179 '"url" field is missing or empty - skipping format, '
2180 'there is an error in extractor')
2182 if isinstance(url
, bytes):
2183 sanitize_string_field(f
, 'url')
2186 # Filter out malformed formats for better extraction robustness
2187 formats
= list(filter(is_wellformed
, formats
))
2191 # We check that all the formats have the format and format_id fields
2192 for i
, format
in enumerate(formats
):
2193 sanitize_string_field(format
, 'format_id')
2194 sanitize_numeric_fields(format
)
2195 format
['url'] = sanitize_url(format
['url'])
2196 if not format
.get('format_id'):
2197 format
['format_id'] = compat_str(i
)
2199 # Sanitize format_id from characters used in format selector expression
2200 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2201 format_id
= format
['format_id']
2202 if format_id
not in formats_dict
:
2203 formats_dict
[format_id
] = []
2204 formats_dict
[format_id
].append(format
)
2206 # Make sure all formats have unique format_id
2207 for format_id
, ambiguous_formats
in formats_dict
.items():
2208 if len(ambiguous_formats
) > 1:
2209 for i
, format
in enumerate(ambiguous_formats
):
2210 format
['format_id'] = '%s-%d' % (format_id
, i
)
2212 for i
, format
in enumerate(formats
):
2213 if format
.get('format') is None:
2214 format
['format'] = '{id} - {res}{note}'.format(
2215 id=format
['format_id'],
2216 res
=self
.format_resolution(format
),
2217 note
=format_field(format
, 'format_note', ' (%s)'),
2219 # Automatically determine file extension if missing
2220 if format
.get('ext') is None:
2221 format
['ext'] = determine_ext(format
['url']).lower()
2222 # Automatically determine protocol if missing (useful for format
2223 # selection purposes)
2224 if format
.get('protocol') is None:
2225 format
['protocol'] = determine_protocol(format
)
2226 # Add HTTP headers, so that external programs can use them from the
2228 full_format_info
= info_dict
.copy()
2229 full_format_info
.update(format
)
2230 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2231 # Remove private housekeeping stuff
2232 if '__x_forwarded_for_ip' in info_dict
:
2233 del info_dict
['__x_forwarded_for_ip']
2235 # TODO Central sorting goes here
2237 if not formats
or formats
[0] is not info_dict
:
2238 # only set the 'formats' fields if the original info_dict list them
2239 # otherwise we end up with a circular reference, the first (and unique)
2240 # element in the 'formats' field in info_dict is info_dict itself,
2241 # which can't be exported to json
2242 info_dict
['formats'] = formats
2244 info_dict
, _
= self
.pre_process(info_dict
)
2246 if self
.params
.get('list_thumbnails'):
2247 self
.list_thumbnails(info_dict
)
2248 if self
.params
.get('listformats'):
2249 if not info_dict
.get('formats') and not info_dict
.get('url'):
2250 self
.to_screen('%s has no formats' % info_dict
['id'])
2252 self
.list_formats(info_dict
)
2253 if self
.params
.get('listsubtitles'):
2254 if 'automatic_captions' in info_dict
:
2255 self
.list_subtitles(
2256 info_dict
['id'], automatic_captions
, 'automatic captions')
2257 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2258 list_only
= self
.params
.get('simulate') is None and (
2259 self
.params
.get('list_thumbnails') or self
.params
.get('listformats') or self
.params
.get('listsubtitles'))
2261 # Without this printing, -F --print-json will not work
2262 self
.__forced
_printings
(info_dict
, self
.prepare_filename(info_dict
), incomplete
=True)
2265 format_selector
= self
.format_selector
2266 if format_selector
is None:
2267 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2268 self
.write_debug('Default format spec: %s' % req_format
)
2269 format_selector
= self
.build_format_selector(req_format
)
2271 # While in format selection we may need to have an access to the original
2272 # format set in order to calculate some metrics or do some processing.
2273 # For now we need to be able to guess whether original formats provided
2274 # by extractor are incomplete or not (i.e. whether extractor provides only
2275 # video-only or audio-only formats) for proper formats selection for
2276 # extractors with such incomplete formats (see
2277 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2278 # Since formats may be filtered during format selection and may not match
2279 # the original formats the results may be incorrect. Thus original formats
2280 # or pre-calculated metrics should be passed to format selection routines
2282 # We will pass a context object containing all necessary additional data
2283 # instead of just formats.
2284 # This fixes incorrect format selection issue (see
2285 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2286 incomplete_formats
= (
2287 # All formats are video-only or
2288 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2289 # all formats are audio-only
2290 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2294 'incomplete_formats': incomplete_formats
,
2297 formats_to_download
= list(format_selector(ctx
))
2298 if not formats_to_download
:
2299 if not self
.params
.get('ignore_no_formats_error'):
2300 raise ExtractorError('Requested format is not available', expected
=True)
2302 self
.report_warning('Requested format is not available')
2303 # Process what we can, even without any available formats.
2304 self
.process_info(dict(info_dict
))
2307 '[info] %s: Downloading %d format(s): %s' % (
2308 info_dict
['id'], len(formats_to_download
),
2309 ", ".join([f
['format_id'] for f
in formats_to_download
])))
2310 for fmt
in formats_to_download
:
2311 new_info
= dict(info_dict
)
2312 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2313 new_info
['__original_infodict'] = info_dict
2314 new_info
.update(fmt
)
2315 self
.process_info(new_info
)
2316 # We update the info dict with the best quality format (backwards compatibility)
2317 if formats_to_download
:
2318 info_dict
.update(formats_to_download
[-1])
2321 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2322 """Select the requested subtitles and their format"""
2324 if normal_subtitles
and self
.params
.get('writesubtitles'):
2325 available_subs
.update(normal_subtitles
)
2326 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2327 for lang
, cap_info
in automatic_captions
.items():
2328 if lang
not in available_subs
:
2329 available_subs
[lang
] = cap_info
2331 if (not self
.params
.get('writesubtitles') and not
2332 self
.params
.get('writeautomaticsub') or not
2336 all_sub_langs
= available_subs
.keys()
2337 if self
.params
.get('allsubtitles', False):
2338 requested_langs
= all_sub_langs
2339 elif self
.params
.get('subtitleslangs', False):
2340 requested_langs
= set()
2341 for lang
in self
.params
.get('subtitleslangs'):
2343 requested_langs
.update(all_sub_langs
)
2345 discard
= lang
[0] == '-'
2348 current_langs
= filter(re
.compile(lang
+ '$').match
, all_sub_langs
)
2350 for lang
in current_langs
:
2351 requested_langs
.discard(lang
)
2353 requested_langs
.update(current_langs
)
2354 elif 'en' in available_subs
:
2355 requested_langs
= ['en']
2357 requested_langs
= [list(all_sub_langs
)[0]]
2359 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2361 formats_query
= self
.params
.get('subtitlesformat', 'best')
2362 formats_preference
= formats_query
.split('/') if formats_query
else []
2364 for lang
in requested_langs
:
2365 formats
= available_subs
.get(lang
)
2367 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2369 for ext
in formats_preference
:
2373 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2379 self
.report_warning(
2380 'No subtitle format found matching "%s" for language %s, '
2381 'using %s' % (formats_query
, lang
, f
['ext']))
2385 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2386 def print_mandatory(field
, actual_field
=None):
2387 if actual_field
is None:
2388 actual_field
= field
2389 if (self
.params
.get('force%s' % field
, False)
2390 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2391 self
.to_stdout(info_dict
[actual_field
])
2393 def print_optional(field
):
2394 if (self
.params
.get('force%s' % field
, False)
2395 and info_dict
.get(field
) is not None):
2396 self
.to_stdout(info_dict
[field
])
2398 info_dict
= info_dict
.copy()
2399 if filename
is not None:
2400 info_dict
['filename'] = filename
2401 if info_dict
.get('requested_formats') is not None:
2402 # For RTMP URLs, also include the playpath
2403 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2404 elif 'url' in info_dict
:
2405 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2407 if self
.params
.get('forceprint') or self
.params
.get('forcejson'):
2408 self
.post_extract(info_dict
)
2409 for tmpl
in self
.params
.get('forceprint', []):
2410 if re
.match(r
'\w+$', tmpl
):
2411 tmpl
= '%({})s'.format(tmpl
)
2412 tmpl
, info_copy
= self
.prepare_outtmpl(tmpl
, info_dict
)
2413 self
.to_stdout(self
.escape_outtmpl(tmpl
) % info_copy
)
2415 print_mandatory('title')
2416 print_mandatory('id')
2417 print_mandatory('url', 'urls')
2418 print_optional('thumbnail')
2419 print_optional('description')
2420 print_optional('filename')
2421 if self
.params
.get('forceduration') and info_dict
.get('duration') is not None:
2422 self
.to_stdout(formatSeconds(info_dict
['duration']))
2423 print_mandatory('format')
2425 if self
.params
.get('forcejson'):
2426 self
.to_stdout(json
.dumps(self
.sanitize_info(info_dict
)))
2428 def dl(self
, name
, info
, subtitle
=False, test
=False):
2429 if not info
.get('url'):
2430 self
.raise_no_formats(info
.get('__has_drm'), forced
=True)
2433 verbose
= self
.params
.get('verbose')
2436 'quiet': not verbose
,
2438 'noprogress': not verbose
,
2440 'skip_unavailable_fragments': False,
2441 'keep_fragments': False,
2443 '_no_ytdl_file': True,
2446 params
= self
.params
2447 fd
= get_suitable_downloader(info
, params
, to_stdout
=(name
== '-'))(self
, params
)
2449 for ph
in self
._progress
_hooks
:
2450 fd
.add_progress_hook(ph
)
2451 urls
= '", "'.join([f
['url'] for f
in info
.get('requested_formats', [])] or [info
['url']])
2452 self
.write_debug('Invoking downloader on "%s"' % urls
)
2453 new_info
= dict(info
)
2454 if new_info
.get('http_headers') is None:
2455 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2456 return fd
.download(name
, new_info
, subtitle
)
2458 def process_info(self
, info_dict
):
2459 """Process a single resolved IE result."""
2461 assert info_dict
.get('_type', 'video') == 'video'
2463 max_downloads
= self
.params
.get('max_downloads')
2464 if max_downloads
is not None:
2465 if self
._num
_downloads
>= int(max_downloads
):
2466 raise MaxDownloadsReached()
2468 # TODO: backward compatibility, to be removed
2469 info_dict
['fulltitle'] = info_dict
['title']
2471 if 'format' not in info_dict
and 'ext' in info_dict
:
2472 info_dict
['format'] = info_dict
['ext']
2474 if self
._match
_entry
(info_dict
) is not None:
2477 self
.post_extract(info_dict
)
2478 self
._num
_downloads
+= 1
2480 # info_dict['_filename'] needs to be set for backward compatibility
2481 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2482 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2486 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=('format' not in info_dict
))
2488 if self
.params
.get('simulate'):
2489 if self
.params
.get('force_write_download_archive', False):
2490 self
.record_download_archive(info_dict
)
2492 # Do nothing else if in simulate mode
2495 if full_filename
is None:
2498 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2500 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2503 if self
.params
.get('writedescription', False):
2504 descfn
= self
.prepare_filename(info_dict
, 'description')
2505 if not self
._ensure
_dir
_exists
(encodeFilename(descfn
)):
2507 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(descfn
)):
2508 self
.to_screen('[info] Video description is already present')
2509 elif info_dict
.get('description') is None:
2510 self
.report_warning('There\'s no description to write.')
2513 self
.to_screen('[info] Writing video description to: ' + descfn
)
2514 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
2515 descfile
.write(info_dict
['description'])
2516 except (OSError, IOError):
2517 self
.report_error('Cannot write description file ' + descfn
)
2520 if self
.params
.get('writeannotations', False):
2521 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2522 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2524 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2525 self
.to_screen('[info] Video annotations are already present')
2526 elif not info_dict
.get('annotations'):
2527 self
.report_warning('There are no annotations to write.')
2530 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2531 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2532 annofile
.write(info_dict
['annotations'])
2533 except (KeyError, TypeError):
2534 self
.report_warning('There are no annotations to write.')
2535 except (OSError, IOError):
2536 self
.report_error('Cannot write annotations file: ' + annofn
)
2539 subtitles_are_requested
= any([self
.params
.get('writesubtitles', False),
2540 self
.params
.get('writeautomaticsub')])
2542 if subtitles_are_requested
and info_dict
.get('requested_subtitles'):
2543 # subtitles download errors are already managed as troubles in relevant IE
2544 # that way it will silently go on when used with unsupporting IE
2545 subtitles
= info_dict
['requested_subtitles']
2546 # ie = self.get_info_extractor(info_dict['extractor_key'])
2547 for sub_lang
, sub_info
in subtitles
.items():
2548 sub_format
= sub_info
['ext']
2549 sub_filename
= subtitles_filename(temp_filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
2550 sub_filename_final
= subtitles_filename(
2551 self
.prepare_filename(info_dict
, 'subtitle'), sub_lang
, sub_format
, info_dict
.get('ext'))
2552 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(sub_filename
)):
2553 self
.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang
, sub_format
))
2554 sub_info
['filepath'] = sub_filename
2555 files_to_move
[sub_filename
] = sub_filename_final
2557 self
.to_screen('[info] Writing video subtitles to: ' + sub_filename
)
2558 if sub_info
.get('data') is not None:
2560 # Use newline='' to prevent conversion of newline characters
2561 # See https://github.com/ytdl-org/youtube-dl/issues/10268
2562 with io
.open(encodeFilename(sub_filename
), 'w', encoding
='utf-8', newline
='') as subfile
:
2563 subfile
.write(sub_info
['data'])
2564 sub_info
['filepath'] = sub_filename
2565 files_to_move
[sub_filename
] = sub_filename_final
2566 except (OSError, IOError):
2567 self
.report_error('Cannot write subtitles file ' + sub_filename
)
2571 self
.dl(sub_filename
, sub_info
.copy(), subtitle
=True)
2572 sub_info
['filepath'] = sub_filename
2573 files_to_move
[sub_filename
] = sub_filename_final
2574 except (ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
2575 self
.report_warning('Unable to download subtitle for "%s": %s' %
2576 (sub_lang
, error_to_compat_str(err
)))
2579 if self
.params
.get('writeinfojson', False):
2580 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2581 if not self
._ensure
_dir
_exists
(encodeFilename(infofn
)):
2583 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(infofn
)):
2584 self
.to_screen('[info] Video metadata is already present')
2586 self
.to_screen('[info] Writing video metadata as JSON to: ' + infofn
)
2588 write_json_file(self
.sanitize_info(info_dict
, self
.params
.get('clean_infojson', True)), infofn
)
2589 except (OSError, IOError):
2590 self
.report_error('Cannot write video metadata to JSON file ' + infofn
)
2592 info_dict
['__infojson_filename'] = infofn
2594 for thumb_ext
in self
._write
_thumbnails
(info_dict
, temp_filename
):
2595 thumb_filename_temp
= replace_extension(temp_filename
, thumb_ext
, info_dict
.get('ext'))
2596 thumb_filename
= replace_extension(
2597 self
.prepare_filename(info_dict
, 'thumbnail'), thumb_ext
, info_dict
.get('ext'))
2598 files_to_move
[thumb_filename_temp
] = thumb_filename
2600 # Write internet shortcut files
2601 url_link
= webloc_link
= desktop_link
= False
2602 if self
.params
.get('writelink', False):
2603 if sys
.platform
== "darwin": # macOS.
2605 elif sys
.platform
.startswith("linux"):
2607 else: # if sys.platform in ['win32', 'cygwin']:
2609 if self
.params
.get('writeurllink', False):
2611 if self
.params
.get('writewebloclink', False):
2613 if self
.params
.get('writedesktoplink', False):
2616 if url_link
or webloc_link
or desktop_link
:
2617 if 'webpage_url' not in info_dict
:
2618 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2620 ascii_url
= iri_to_uri(info_dict
['webpage_url'])
2622 def _write_link_file(extension
, template
, newline
, embed_filename
):
2623 linkfn
= replace_extension(full_filename
, extension
, info_dict
.get('ext'))
2624 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2625 self
.to_screen('[info] Internet shortcut is already present')
2628 self
.to_screen('[info] Writing internet shortcut to: ' + linkfn
)
2629 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8', newline
=newline
) as linkfile
:
2630 template_vars
= {'url': ascii_url}
2632 template_vars
['filename'] = linkfn
[:-(len(extension
) + 1)]
2633 linkfile
.write(template
% template_vars
)
2634 except (OSError, IOError):
2635 self
.report_error('Cannot write internet shortcut ' + linkfn
)
2640 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE
, '\r\n', embed_filename
=False):
2643 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE
, '\n', embed_filename
=False):
2646 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE
, '\n', embed_filename
=True):
2650 info_dict
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2651 except PostProcessingError
as err
:
2652 self
.report_error('Preprocessing: %s' % str(err
))
2655 must_record_download_archive
= False
2656 if self
.params
.get('skip_download', False):
2657 info_dict
['filepath'] = temp_filename
2658 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2659 info_dict
['__files_to_move'] = files_to_move
2660 info_dict
= self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
)
2663 info_dict
.setdefault('__postprocessors', [])
2666 def existing_file(*filepaths
):
2667 ext
= info_dict
.get('ext')
2668 final_ext
= self
.params
.get('final_ext', ext
)
2670 for file in orderedSet(filepaths
):
2671 if final_ext
!= ext
:
2672 converted
= replace_extension(file, final_ext
, ext
)
2673 if os
.path
.exists(encodeFilename(converted
)):
2674 existing_files
.append(converted
)
2675 if os
.path
.exists(encodeFilename(file)):
2676 existing_files
.append(file)
2678 if not existing_files
or self
.params
.get('overwrites', False):
2679 for file in orderedSet(existing_files
):
2680 self
.report_file_delete(file)
2681 os
.remove(encodeFilename(file))
2684 info_dict
['ext'] = os
.path
.splitext(existing_files
[0])[1][1:]
2685 return existing_files
[0]
2688 if info_dict
.get('requested_formats') is not None:
2690 def compatible_formats(formats
):
2691 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2692 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2693 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2694 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2698 exts
= set(format
.get('ext') for format
in formats
)
2700 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2703 for ext_sets
in COMPATIBLE_EXTS
:
2704 if ext_sets
.issuperset(exts
):
2706 # TODO: Check acodec/vcodec
2709 requested_formats
= info_dict
['requested_formats']
2710 old_ext
= info_dict
['ext']
2711 if self
.params
.get('merge_output_format') is None and not compatible_formats(requested_formats
):
2712 info_dict
['ext'] = 'mkv'
2713 self
.report_warning(
2714 'Requested formats are incompatible for merge and will be merged into mkv.')
2715 new_ext
= info_dict
['ext']
2717 def correct_ext(filename
, ext
=new_ext
):
2720 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2722 os
.path
.splitext(filename
)[0]
2723 if filename_real_ext
in (old_ext
, new_ext
)
2725 return '%s.%s' % (filename_wo_ext
, ext
)
2727 # Ensure filename always has a correct extension for successful merge
2728 full_filename
= correct_ext(full_filename
)
2729 temp_filename
= correct_ext(temp_filename
)
2730 dl_filename
= existing_file(full_filename
, temp_filename
)
2731 info_dict
['__real_download'] = False
2733 _protocols
= set(determine_protocol(f
) for f
in requested_formats
)
2734 if len(_protocols
) == 1: # All requested formats have same protocol
2735 info_dict
['protocol'] = _protocols
.pop()
2736 directly_mergable
= FFmpegFD
.can_merge_formats(info_dict
)
2737 if dl_filename
is not None:
2738 self
.report_file_already_downloaded(dl_filename
)
2739 elif (directly_mergable
and get_suitable_downloader(
2740 info_dict
, self
.params
, to_stdout
=(temp_filename
== '-')) == FFmpegFD
):
2741 info_dict
['url'] = '\n'.join(f
['url'] for f
in requested_formats
)
2742 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2743 info_dict
['__real_download'] = real_download
2746 merger
= FFmpegMergerPP(self
)
2747 if self
.params
.get('allow_unplayable_formats'):
2748 self
.report_warning(
2749 'You have requested merging of multiple formats '
2750 'while also allowing unplayable formats to be downloaded. '
2751 'The formats won\'t be merged to prevent data corruption.')
2752 elif not merger
.available
:
2753 self
.report_warning(
2754 'You have requested merging of multiple formats but ffmpeg is not installed. '
2755 'The formats won\'t be merged.')
2757 if temp_filename
== '-':
2758 reason
= ('using a downloader other than ffmpeg' if directly_mergable
2759 else 'but the formats are incompatible for simultaneous download' if merger
.available
2760 else 'but ffmpeg is not installed')
2761 self
.report_warning(
2762 f
'You have requested downloading multiple formats to stdout {reason}. '
2763 'The formats will be streamed one after the other')
2764 fname
= temp_filename
2765 for f
in requested_formats
:
2766 new_info
= dict(info_dict
)
2767 del new_info
['requested_formats']
2769 if temp_filename
!= '-':
2770 fname
= prepend_extension(
2771 correct_ext(temp_filename
, new_info
['ext']),
2772 'f%s' % f
['format_id'], new_info
['ext'])
2773 if not self
._ensure
_dir
_exists
(fname
):
2775 downloaded
.append(fname
)
2776 partial_success
, real_download
= self
.dl(fname
, new_info
)
2777 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
2778 success
= success
and partial_success
2779 if merger
.available
and not self
.params
.get('allow_unplayable_formats'):
2780 info_dict
['__postprocessors'].append(merger
)
2781 info_dict
['__files_to_merge'] = downloaded
2782 # Even if there were no downloads, it is being merged only now
2783 info_dict
['__real_download'] = True
2785 for file in downloaded
:
2786 files_to_move
[file] = None
2788 # Just a single file
2789 dl_filename
= existing_file(full_filename
, temp_filename
)
2790 if dl_filename
is None or dl_filename
== temp_filename
:
2791 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
2792 # So we should try to resume the download
2793 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2794 info_dict
['__real_download'] = real_download
2796 self
.report_file_already_downloaded(dl_filename
)
2798 dl_filename
= dl_filename
or temp_filename
2799 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2801 except network_exceptions
as err
:
2802 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
2804 except (OSError, IOError) as err
:
2805 raise UnavailableVideoError(err
)
2806 except (ContentTooShortError
, ) as err
:
2807 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
2810 if success
and full_filename
!= '-':
2814 fixup_policy
= self
.params
.get('fixup')
2815 vid
= info_dict
['id']
2817 if fixup_policy
in ('ignore', 'never'):
2819 elif fixup_policy
== 'warn':
2821 elif fixup_policy
!= 'force':
2822 assert fixup_policy
in ('detect_or_warn', None)
2823 if not info_dict
.get('__real_download'):
2826 def ffmpeg_fixup(cndn
, msg
, cls
):
2830 self
.report_warning(f
'{vid}: {msg}')
2834 info_dict
['__postprocessors'].append(pp
)
2836 self
.report_warning(f
'{vid}: {msg}. Install ffmpeg to fix this automatically')
2838 stretched_ratio
= info_dict
.get('stretched_ratio')
2840 stretched_ratio
not in (1, None),
2841 f
'Non-uniform pixel ratio {stretched_ratio}',
2842 FFmpegFixupStretchedPP
)
2845 (info_dict
.get('requested_formats') is None
2846 and info_dict
.get('container') == 'm4a_dash'
2847 and info_dict
.get('ext') == 'm4a'),
2848 'writing DASH m4a. Only some players support this container',
2851 downloader
= (get_suitable_downloader(info_dict
, self
.params
).__name
__
2852 if 'protocol' in info_dict
else None)
2853 ffmpeg_fixup(downloader
== 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP
)
2854 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP
)
2855 ffmpeg_fixup(downloader
== 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP
)
2859 info_dict
= self
.post_process(dl_filename
, info_dict
, files_to_move
)
2860 except PostProcessingError
as err
:
2861 self
.report_error('Postprocessing: %s' % str(err
))
2864 for ph
in self
._post
_hooks
:
2865 ph(info_dict
['filepath'])
2866 except Exception as err
:
2867 self
.report_error('post hooks: %s' % str(err
))
2869 must_record_download_archive
= True
2871 if must_record_download_archive
or self
.params
.get('force_write_download_archive', False):
2872 self
.record_download_archive(info_dict
)
2873 max_downloads
= self
.params
.get('max_downloads')
2874 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
2875 raise MaxDownloadsReached()
2877 def download(self
, url_list
):
2878 """Download a given list of URLs."""
2879 outtmpl
= self
.outtmpl_dict
['default']
2880 if (len(url_list
) > 1
2882 and '%' not in outtmpl
2883 and self
.params
.get('max_downloads') != 1):
2884 raise SameFileError(outtmpl
)
2886 for url
in url_list
:
2888 # It also downloads the videos
2889 res
= self
.extract_info(
2890 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
2891 except UnavailableVideoError
:
2892 self
.report_error('unable to download video')
2893 except MaxDownloadsReached
:
2894 self
.to_screen('[info] Maximum number of downloads reached')
2896 except ExistingVideoReached
:
2897 self
.to_screen('[info] Encountered a video that is already in the archive, stopping due to --break-on-existing')
2899 except RejectedVideoReached
:
2900 self
.to_screen('[info] Encountered a video that did not match filter, stopping due to --break-on-reject')
2903 if self
.params
.get('dump_single_json', False):
2904 self
.post_extract(res
)
2905 self
.to_stdout(json
.dumps(self
.sanitize_info(res
)))
2907 return self
._download
_retcode
2909 def download_with_info_file(self
, info_filename
):
2910 with contextlib
.closing(fileinput
.FileInput(
2911 [info_filename
], mode
='r',
2912 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
2913 # FileInput doesn't have a read method, we can't call json.load
2914 info
= self
.sanitize_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
2916 self
.process_ie_result(info
, download
=True)
2917 except (DownloadError
, EntryNotInPlaylist
, ThrottledDownload
):
2918 webpage_url
= info
.get('webpage_url')
2919 if webpage_url
is not None:
2920 self
.report_warning('The info failed to download, trying with "%s"' % webpage_url
)
2921 return self
.download([webpage_url
])
2924 return self
._download
_retcode
2927 def sanitize_info(info_dict
, remove_private_keys
=False):
2928 ''' Sanitize the infodict for converting to json '''
2929 if info_dict
is None:
2931 info_dict
.setdefault('epoch', int(time
.time()))
2932 remove_keys
= {'__original_infodict'}
# Always remove this since this may contain a copy of the entire dict
2933 keep_keys
= ['_type'], # Always keep this to facilitate load-info-json
2934 if remove_private_keys
:
2936 'requested_formats', 'requested_subtitles', 'requested_entries',
2937 'filepath', 'entries', 'original_url', 'playlist_autonumber',
2939 empty_values
= (None, {}, [], set(), tuple())
2940 reject
= lambda k
, v
: k
not in keep_keys
and (
2941 k
.startswith('_') or k
in remove_keys
or v
in empty_values
)
2943 reject
= lambda k
, v
: k
in remove_keys
2944 filter_fn
= lambda obj
: (
2945 list(map(filter_fn
, obj
)) if isinstance(obj
, (LazyList
, list, tuple, set))
2946 else obj
if not isinstance(obj
, dict)
2947 else dict((k
, filter_fn(v
)) for k
, v
in obj
.items() if not reject(k
, v
)))
2948 return filter_fn(info_dict
)
2951 def filter_requested_info(info_dict
, actually_filter
=True):
2952 ''' Alias of sanitize_info for backward compatibility '''
2953 return YoutubeDL
.sanitize_info(info_dict
, actually_filter
)
2955 def run_pp(self
, pp
, infodict
):
2956 files_to_delete
= []
2957 if '__files_to_move' not in infodict
:
2958 infodict
['__files_to_move'] = {}
2959 files_to_delete
, infodict
= pp
.run(infodict
)
2960 if not files_to_delete
:
2963 if self
.params
.get('keepvideo', False):
2964 for f
in files_to_delete
:
2965 infodict
['__files_to_move'].setdefault(f
, '')
2967 for old_filename
in set(files_to_delete
):
2968 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
2970 os
.remove(encodeFilename(old_filename
))
2971 except (IOError, OSError):
2972 self
.report_warning('Unable to remove downloaded original file')
2973 if old_filename
in infodict
['__files_to_move']:
2974 del infodict
['__files_to_move'][old_filename
]
2978 def post_extract(info_dict
):
2979 def actual_post_extract(info_dict
):
2980 if info_dict
.get('_type') in ('playlist', 'multi_video'):
2981 for video_dict
in info_dict
.get('entries', {}):
2982 actual_post_extract(video_dict
or {})
2985 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
2986 extra
= post_extractor().items()
2987 info_dict
.update(extra
)
2988 info_dict
.pop('__post_extractor', None)
2990 original_infodict
= info_dict
.get('__original_infodict') or {}
2991 original_infodict
.update(extra
)
2992 original_infodict
.pop('__post_extractor', None)
2994 actual_post_extract(info_dict
or {})
2996 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
2997 info
= dict(ie_info
)
2998 info
['__files_to_move'] = files_to_move
or {}
2999 for pp
in self
._pps
[key
]:
3000 info
= self
.run_pp(pp
, info
)
3001 return info
, info
.pop('__files_to_move', None)
3003 def post_process(self
, filename
, ie_info
, files_to_move
=None):
3004 """Run all the postprocessors on the given file."""
3005 info
= dict(ie_info
)
3006 info
['filepath'] = filename
3007 info
['__files_to_move'] = files_to_move
or {}
3009 for pp
in ie_info
.get('__postprocessors', []) + self
._pps
['post_process']:
3010 info
= self
.run_pp(pp
, info
)
3011 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
3012 del info
['__files_to_move']
3013 for pp
in self
._pps
['after_move']:
3014 info
= self
.run_pp(pp
, info
)
3017 def _make_archive_id(self
, info_dict
):
3018 video_id
= info_dict
.get('id')
3021 # Future-proof against any change in case
3022 # and backwards compatibility with prior versions
3023 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
3024 if extractor
is None:
3025 url
= str_or_none(info_dict
.get('url'))
3028 # Try to find matching extractor for the URL and take its ie_key
3029 for ie
in self
._ies
:
3030 if ie
.suitable(url
):
3031 extractor
= ie
.ie_key()
3035 return '%s %s' % (extractor
.lower(), video_id
)
3037 def in_download_archive(self
, info_dict
):
3038 fn
= self
.params
.get('download_archive')
3042 vid_id
= self
._make
_archive
_id
(info_dict
)
3044 return False # Incomplete video information
3046 return vid_id
in self
.archive
3048 def record_download_archive(self
, info_dict
):
3049 fn
= self
.params
.get('download_archive')
3052 vid_id
= self
._make
_archive
_id
(info_dict
)
3054 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
3055 archive_file
.write(vid_id
+ '\n')
3056 self
.archive
.add(vid_id
)
3059 def format_resolution(format
, default
='unknown'):
3060 if format
.get('vcodec') == 'none':
3061 if format
.get('acodec') == 'none':
3064 if format
.get('resolution') is not None:
3065 return format
['resolution']
3066 if format
.get('width') and format
.get('height'):
3067 res
= '%dx%d' % (format
['width'], format
['height'])
3068 elif format
.get('height'):
3069 res
= '%sp' % format
['height']
3070 elif format
.get('width'):
3071 res
= '%dx?' % format
['width']
3076 def _format_note(self
, fdict
):
3078 if fdict
.get('ext') in ['f4f', 'f4m']:
3079 res
+= '(unsupported) '
3080 if fdict
.get('language'):
3083 res
+= '[%s] ' % fdict
['language']
3084 if fdict
.get('format_note') is not None:
3085 res
+= fdict
['format_note'] + ' '
3086 if fdict
.get('tbr') is not None:
3087 res
+= '%4dk ' % fdict
['tbr']
3088 if fdict
.get('container') is not None:
3091 res
+= '%s container' % fdict
['container']
3092 if (fdict
.get('vcodec') is not None
3093 and fdict
.get('vcodec') != 'none'):
3096 res
+= fdict
['vcodec']
3097 if fdict
.get('vbr') is not None:
3099 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
3101 if fdict
.get('vbr') is not None:
3102 res
+= '%4dk' % fdict
['vbr']
3103 if fdict
.get('fps') is not None:
3106 res
+= '%sfps' % fdict
['fps']
3107 if fdict
.get('acodec') is not None:
3110 if fdict
['acodec'] == 'none':
3113 res
+= '%-5s' % fdict
['acodec']
3114 elif fdict
.get('abr') is not None:
3118 if fdict
.get('abr') is not None:
3119 res
+= '@%3dk' % fdict
['abr']
3120 if fdict
.get('asr') is not None:
3121 res
+= ' (%5dHz)' % fdict
['asr']
3122 if fdict
.get('filesize') is not None:
3125 res
+= format_bytes(fdict
['filesize'])
3126 elif fdict
.get('filesize_approx') is not None:
3129 res
+= '~' + format_bytes(fdict
['filesize_approx'])
3132 def list_formats(self
, info_dict
):
3133 formats
= info_dict
.get('formats', [info_dict
])
3135 'list-formats' not in self
.params
.get('compat_opts', [])
3136 and self
.params
.get('listformats_table', True) is not False)
3140 format_field(f
, 'format_id'),
3141 format_field(f
, 'ext'),
3142 self
.format_resolution(f
),
3143 format_field(f
, 'fps', '%d'),
3145 format_field(f
, 'filesize', ' %s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~%s', func
=format_bytes
),
3146 format_field(f
, 'tbr', '%4dk'),
3147 shorten_protocol_name(f
.get('protocol', '').replace("native", "n")),
3149 format_field(f
, 'vcodec', default
='unknown').replace('none', ''),
3150 format_field(f
, 'vbr', '%4dk'),
3151 format_field(f
, 'acodec', default
='unknown').replace('none', ''),
3152 format_field(f
, 'abr', '%3dk'),
3153 format_field(f
, 'asr', '%5dHz'),
3154 ', '.join(filter(None, (
3155 'UNSUPPORTED' if f
.get('ext') in ('f4f', 'f4m') else '',
3156 format_field(f
, 'language', '[%s]'),
3157 format_field(f
, 'format_note'),
3158 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
3160 ] for f
in formats
if f
.get('preference') is None or f
['preference'] >= -1000]
3161 header_line
= ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3162 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO']
3166 format_field(f
, 'format_id'),
3167 format_field(f
, 'ext'),
3168 self
.format_resolution(f
),
3169 self
._format
_note
(f
)]
3171 if f
.get('preference') is None or f
['preference'] >= -1000]
3172 header_line
= ['format code', 'extension', 'resolution', 'note']
3175 '[info] Available formats for %s:' % info_dict
['id'])
3176 self
.to_stdout(render_table(
3177 header_line
, table
, delim
=new_format
, extraGap
=(0 if new_format
else 1), hideEmpty
=new_format
))
3179 def list_thumbnails(self
, info_dict
):
3180 thumbnails
= list(info_dict
.get('thumbnails'))
3182 self
.to_screen('[info] No thumbnails present for %s' % info_dict
['id'])
3186 '[info] Thumbnails for %s:' % info_dict
['id'])
3187 self
.to_stdout(render_table(
3188 ['ID', 'width', 'height', 'URL'],
3189 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
]))
3191 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3193 self
.to_screen('%s has no %s' % (video_id
, name
))
3196 'Available %s for %s:' % (name
, video_id
))
3198 def _row(lang
, formats
):
3199 exts
, names
= zip(*((f
['ext'], f
.get('name') or 'unknown') for f
in reversed(formats
)))
3200 if len(set(names
)) == 1:
3201 names
= [] if names
[0] == 'unknown' else names
[:1]
3202 return [lang
, ', '.join(names
), ', '.join(exts
)]
3204 self
.to_stdout(render_table(
3205 ['Language', 'Name', 'Formats'],
3206 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3209 def urlopen(self
, req
):
3210 """ Start an HTTP download """
3211 if isinstance(req
, compat_basestring
):
3212 req
= sanitized_Request(req
)
3213 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3215 def print_debug_header(self
):
3216 if not self
.params
.get('verbose'):
3219 stdout_encoding
= getattr(
3220 sys
.stdout
, 'encoding', 'missing (%s)' % type(sys
.stdout
).__name
__)
3222 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3223 locale
.getpreferredencoding(),
3224 sys
.getfilesystemencoding(),
3226 self
.get_encoding()))
3227 write_string(encoding_str
, encoding
=None)
3230 '(exe)' if hasattr(sys
, 'frozen')
3231 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter
)
3232 else '(source)' if os
.path
.basename(sys
.argv
[0]) == '__main__.py'
3234 self
._write
_string
('[debug] yt-dlp version %s %s\n' % (__version__
, source
))
3236 self
._write
_string
('[debug] Lazy loading extractors enabled\n')
3239 '[debug] Plugin Extractors: %s\n' % [ie
.ie_key() for ie
in _PLUGIN_CLASSES
])
3240 if self
.params
.get('compat_opts'):
3242 '[debug] Compatibility options: %s\n' % ', '.join(self
.params
.get('compat_opts')))
3244 sp
= subprocess
.Popen(
3245 ['git', 'rev-parse', '--short', 'HEAD'],
3246 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3247 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3248 out
, err
= process_communicate_or_kill(sp
)
3249 out
= out
.decode().strip()
3250 if re
.match('[0-9a-f]+', out
):
3251 self
._write
_string
('[debug] Git HEAD: %s\n' % out
)
3258 def python_implementation():
3259 impl_name
= platform
.python_implementation()
3260 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3261 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3264 self
._write
_string
('[debug] Python version %s (%s %s) - %s\n' % (
3265 platform
.python_version(),
3266 python_implementation(),
3267 platform
.architecture()[0],
3270 exe_versions
= FFmpegPostProcessor
.get_versions(self
)
3271 exe_versions
['rtmpdump'] = rtmpdump_version()
3272 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3273 exe_str
= ', '.join(
3274 f
'{exe} {v}' for exe
, v
in sorted(exe_versions
.items()) if v
3276 self
._write
_string
('[debug] exe versions: %s\n' % exe_str
)
3278 from .downloader
.fragment
import can_decrypt_frag
3279 from .downloader
.websocket
import has_websockets
3280 from .postprocessor
.embedthumbnail
import has_mutagen
3281 from .cookies
import SQLITE_AVAILABLE
, KEYRING_AVAILABLE
3283 lib_str
= ', '.join(sorted(filter(None, (
3284 can_decrypt_frag
and 'pycryptodome',
3285 has_websockets
and 'websockets',
3286 has_mutagen
and 'mutagen',
3287 SQLITE_AVAILABLE
and 'sqlite',
3288 KEYRING_AVAILABLE
and 'keyring',
3290 self
._write
_string
('[debug] Optional libraries: %s\n' % lib_str
)
3293 for handler
in self
._opener
.handlers
:
3294 if hasattr(handler
, 'proxies'):
3295 proxy_map
.update(handler
.proxies
)
3296 self
._write
_string
('[debug] Proxy map: ' + compat_str(proxy_map
) + '\n')
3298 if self
.params
.get('call_home', False):
3299 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3300 self
._write
_string
('[debug] Public IP address: %s\n' % ipaddr
)
3302 latest_version
= self
.urlopen(
3303 'https://yt-dl.org/latest/version').read().decode('utf-8')
3304 if version_tuple(latest_version
) > version_tuple(__version__
):
3305 self
.report_warning(
3306 'You are using an outdated version (newest version: %s)! '
3307 'See https://yt-dl.org/update if you need help updating.' %
3310 def _setup_opener(self
):
3311 timeout_val
= self
.params
.get('socket_timeout')
3312 self
._socket
_timeout
= 600 if timeout_val
is None else float(timeout_val
)
3314 opts_cookiesfrombrowser
= self
.params
.get('cookiesfrombrowser')
3315 opts_cookiefile
= self
.params
.get('cookiefile')
3316 opts_proxy
= self
.params
.get('proxy')
3318 self
.cookiejar
= load_cookies(opts_cookiefile
, opts_cookiesfrombrowser
, self
)
3320 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3321 if opts_proxy
is not None:
3322 if opts_proxy
== '':
3325 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3327 proxies
= compat_urllib_request
.getproxies()
3328 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3329 if 'http' in proxies
and 'https' not in proxies
:
3330 proxies
['https'] = proxies
['http']
3331 proxy_handler
= PerRequestProxyHandler(proxies
)
3333 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3334 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3335 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3336 redirect_handler
= YoutubeDLRedirectHandler()
3337 data_handler
= compat_urllib_request_DataHandler()
3339 # When passing our own FileHandler instance, build_opener won't add the
3340 # default FileHandler and allows us to disable the file protocol, which
3341 # can be used for malicious purposes (see
3342 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3343 file_handler
= compat_urllib_request
.FileHandler()
3345 def file_open(*args
, **kwargs
):
3346 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3347 file_handler
.file_open
= file_open
3349 opener
= compat_urllib_request
.build_opener(
3350 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3352 # Delete the default user-agent header, which would otherwise apply in
3353 # cases where our custom HTTP handler doesn't come into play
3354 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3355 opener
.addheaders
= []
3356 self
._opener
= opener
3358 def encode(self
, s
):
3359 if isinstance(s
, bytes):
3360 return s
# Already encoded
3363 return s
.encode(self
.get_encoding())
3364 except UnicodeEncodeError as err
:
3365 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3368 def get_encoding(self
):
3369 encoding
= self
.params
.get('encoding')
3370 if encoding
is None:
3371 encoding
= preferredencoding()
3374 def _write_thumbnails(self
, info_dict
, filename
): # return the extensions
3375 write_all
= self
.params
.get('write_all_thumbnails', False)
3377 if write_all
or self
.params
.get('writethumbnail', False):
3378 thumbnails
= info_dict
.get('thumbnails') or []
3379 multiple
= write_all
and len(thumbnails
) > 1
3382 for t
in thumbnails
[::-1]:
3383 thumb_ext
= determine_ext(t
['url'], 'jpg')
3384 suffix
= '%s.' % t
['id'] if multiple
else ''
3385 thumb_display_id
= '%s ' % t
['id'] if multiple
else ''
3386 thumb_filename
= replace_extension(filename
, suffix
+ thumb_ext
, info_dict
.get('ext'))
3388 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(thumb_filename
)):
3389 ret
.append(suffix
+ thumb_ext
)
3390 t
['filepath'] = thumb_filename
3391 self
.to_screen('[%s] %s: Thumbnail %sis already present' %
3392 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
3394 self
.to_screen('[%s] %s: Downloading thumbnail %s ...' %
3395 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
3397 uf
= self
.urlopen(t
['url'])
3398 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3399 shutil
.copyfileobj(uf
, thumbf
)
3400 ret
.append(suffix
+ thumb_ext
)
3401 self
.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3402 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
, thumb_filename
))
3403 t
['filepath'] = thumb_filename
3404 except network_exceptions
as err
:
3405 self
.report_warning('Unable to download thumbnail "%s": %s' %
3406 (t
['url'], error_to_compat_str(err
)))
3407 if ret
and not write_all
: