4 from __future__
import absolute_import
, unicode_literals
28 from string
import ascii_letters
29 from zipimport
import zipimporter
34 compat_get_terminal_size
,
39 compat_tokenize_tokenize
,
41 compat_urllib_request
,
42 compat_urllib_request_DataHandler
,
53 DOT_DESKTOP_LINK_TEMPLATE
,
54 DOT_URL_LINK_TEMPLATE
,
55 DOT_WEBLOC_LINK_TEMPLATE
,
83 PerRequestProxyHandler
,
88 process_communicate_or_kill
,
90 register_socks_protocols
,
105 UnavailableVideoError
,
111 YoutubeDLCookieProcessor
,
113 YoutubeDLRedirectHandler
,
115 from .cache
import Cache
116 from .extractor
import (
117 gen_extractor_classes
,
122 from .extractor
.openload
import PhantomJSwrapper
123 from .downloader
import (
124 get_suitable_downloader
,
125 shorten_protocol_name
127 from .downloader
.rtmp
import rtmpdump_version
128 from .postprocessor
import (
131 FFmpegFixupStretchedPP
,
134 # FFmpegSubtitlesConvertorPP,
136 MoveFilesAfterDownloadPP
,
138 from .version
import __version__
140 if compat_os_name
== 'nt':
144 class YoutubeDL(object):
147 YoutubeDL objects are the ones responsible of downloading the
148 actual video file and writing it to disk if the user has requested
149 it, among some other tasks. In most cases there should be one per
150 program. As, given a video URL, the downloader doesn't know how to
151 extract all the needed information, task that InfoExtractors do, it
152 has to pass the URL to one of them.
154 For this, YoutubeDL objects have a method that allows
155 InfoExtractors to be registered in a given order. When it is passed
156 a URL, the YoutubeDL object handles it to the first InfoExtractor it
157 finds that reports being able to handle it. The InfoExtractor extracts
158 all the information about the video or videos the URL refers to, and
159 YoutubeDL process the extracted information, possibly using a File
160 Downloader to download the video.
162 YoutubeDL objects accept a lot of parameters. In order not to saturate
163 the object constructor with arguments, it receives a dictionary of
164 options instead. These options are available through the params
165 attribute for the InfoExtractors to use. The YoutubeDL also
166 registers itself as the downloader in charge for the InfoExtractors
167 that are added to it, so this is a "mutual registration".
171 username: Username for authentication purposes.
172 password: Password for authentication purposes.
173 videopassword: Password for accessing a video.
174 ap_mso: Adobe Pass multiple-system operator identifier.
175 ap_username: Multiple-system operator account username.
176 ap_password: Multiple-system operator account password.
177 usenetrc: Use netrc for authentication instead.
178 verbose: Print additional info to stdout.
179 quiet: Do not print messages to stdout.
180 no_warnings: Do not print out anything for warnings.
181 forceprint: A list of templates to force print
182 forceurl: Force printing final URL. (Deprecated)
183 forcetitle: Force printing title. (Deprecated)
184 forceid: Force printing ID. (Deprecated)
185 forcethumbnail: Force printing thumbnail URL. (Deprecated)
186 forcedescription: Force printing description. (Deprecated)
187 forcefilename: Force printing final filename. (Deprecated)
188 forceduration: Force printing duration. (Deprecated)
189 forcejson: Force printing info_dict as JSON.
190 dump_single_json: Force printing the info_dict of the whole playlist
191 (or video) as a single JSON line.
192 force_write_download_archive: Force writing download archive regardless
193 of 'skip_download' or 'simulate'.
194 simulate: Do not download the video files.
195 format: Video format code. see "FORMAT SELECTION" for more details.
196 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
197 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
198 extracting metadata even if the video is not actually
199 available for download (experimental)
200 format_sort: How to sort the video formats. see "Sorting Formats"
202 format_sort_force: Force the given format_sort. see "Sorting Formats"
204 allow_multiple_video_streams: Allow multiple video streams to be merged
206 allow_multiple_audio_streams: Allow multiple audio streams to be merged
208 paths: Dictionary of output paths. The allowed keys are 'home'
209 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
210 outtmpl: Dictionary of templates for output names. Allowed keys
211 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
212 A string a also accepted for backward compatibility
213 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
214 restrictfilenames: Do not allow "&" and spaces in file names
215 trim_file_name: Limit length of filename (extension excluded)
216 windowsfilenames: Force the filenames to be windows compatible
217 ignoreerrors: Do not stop on download errors
218 (Default True when running yt-dlp,
219 but False when directly accessing YoutubeDL class)
220 skip_playlist_after_errors: Number of allowed failures until the rest of
221 the playlist is skipped
222 force_generic_extractor: Force downloader to use the generic extractor
223 overwrites: Overwrite all video and metadata files if True,
224 overwrite only non-video files if None
225 and don't overwrite any file if False
226 playliststart: Playlist item to start at.
227 playlistend: Playlist item to end at.
228 playlist_items: Specific indices of playlist to download.
229 playlistreverse: Download playlist items in reverse order.
230 playlistrandom: Download playlist items in random order.
231 matchtitle: Download only matching titles.
232 rejecttitle: Reject downloads for matching titles.
233 logger: Log messages to a logging.Logger instance.
234 logtostderr: Log messages to stderr instead of stdout.
235 writedescription: Write the video description to a .description file
236 writeinfojson: Write the video description to a .info.json file
237 clean_infojson: Remove private fields from the infojson
238 writecomments: Extract video comments. This will not be written to disk
239 unless writeinfojson is also given
240 writeannotations: Write the video annotations to a .annotations.xml file
241 writethumbnail: Write the thumbnail image to a file
242 allow_playlist_files: Whether to write playlists' description, infojson etc
243 also to disk when using the 'write*' options
244 write_all_thumbnails: Write all thumbnail formats to files
245 writelink: Write an internet shortcut file, depending on the
246 current platform (.url/.webloc/.desktop)
247 writeurllink: Write a Windows internet shortcut file (.url)
248 writewebloclink: Write a macOS internet shortcut file (.webloc)
249 writedesktoplink: Write a Linux internet shortcut file (.desktop)
250 writesubtitles: Write the video subtitles to a file
251 writeautomaticsub: Write the automatically generated subtitles to a file
252 allsubtitles: Deprecated - Use subtitlelangs = ['all']
253 Downloads all the subtitles of the video
254 (requires writesubtitles or writeautomaticsub)
255 listsubtitles: Lists all available subtitles for the video
256 subtitlesformat: The format code for subtitles
257 subtitleslangs: List of languages of the subtitles to download (can be regex).
258 The list may contain "all" to refer to all the available
259 subtitles. The language can be prefixed with a "-" to
260 exclude it from the requested languages. Eg: ['all', '-live_chat']
261 keepvideo: Keep the video file after post-processing
262 daterange: A DateRange object, download only if the upload_date is in the range.
263 skip_download: Skip the actual download of the video file
264 cachedir: Location of the cache files in the filesystem.
265 False to disable filesystem cache.
266 noplaylist: Download single video instead of a playlist if in doubt.
267 age_limit: An integer representing the user's age in years.
268 Unsuitable videos for the given age are skipped.
269 min_views: An integer representing the minimum view count the video
270 must have in order to not be skipped.
271 Videos without view count information are always
272 downloaded. None for no limit.
273 max_views: An integer representing the maximum view count.
274 Videos that are more popular than that are not
276 Videos without view count information are always
277 downloaded. None for no limit.
278 download_archive: File name of a file where all downloads are recorded.
279 Videos already present in the file are not downloaded
281 break_on_existing: Stop the download process after attempting to download a
282 file that is in the archive.
283 break_on_reject: Stop the download process when encountering a video that
284 has been filtered out.
285 cookiefile: File name where cookies should be read from and dumped to
286 nocheckcertificate:Do not verify SSL certificates
287 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
288 At the moment, this is only supported by YouTube.
289 proxy: URL of the proxy server to use
290 geo_verification_proxy: URL of the proxy to use for IP address verification
291 on geo-restricted sites.
292 socket_timeout: Time to wait for unresponsive hosts, in seconds
293 bidi_workaround: Work around buggy terminals without bidirectional text
294 support, using fridibi
295 debug_printtraffic:Print out sent and received HTTP traffic
296 include_ads: Download ads as well
297 default_search: Prepend this string if an input url is not valid.
298 'auto' for elaborate guessing
299 encoding: Use this encoding instead of the system-specified.
300 extract_flat: Do not resolve URLs, return the immediate result.
301 Pass in 'in_playlist' to only show this behavior for
303 postprocessors: A list of dictionaries, each with an entry
304 * key: The name of the postprocessor. See
305 yt_dlp/postprocessor/__init__.py for a list.
306 * when: When to run the postprocessor. Can be one of
307 pre_process|before_dl|post_process|after_move.
308 Assumed to be 'post_process' if not given
309 post_hooks: A list of functions that get called as the final step
310 for each video file, after all postprocessors have been
311 called. The filename will be passed as the only argument.
312 progress_hooks: A list of functions that get called on download
313 progress, with a dictionary with the entries
314 * status: One of "downloading", "error", or "finished".
315 Check this first and ignore unknown values.
317 If status is one of "downloading", or "finished", the
318 following properties may also be present:
319 * filename: The final filename (always present)
320 * tmpfilename: The filename we're currently writing to
321 * downloaded_bytes: Bytes on disk
322 * total_bytes: Size of the whole file, None if unknown
323 * total_bytes_estimate: Guess of the eventual file size,
325 * elapsed: The number of seconds since download started.
326 * eta: The estimated time in seconds, None if unknown
327 * speed: The download speed in bytes/second, None if
329 * fragment_index: The counter of the currently
330 downloaded video fragment.
331 * fragment_count: The number of fragments (= individual
332 files that will be merged)
334 Progress hooks are guaranteed to be called at least once
335 (with status "finished") if the download is successful.
336 merge_output_format: Extension to use when merging formats.
337 final_ext: Expected final extension; used to detect when the file was
338 already downloaded and converted. "merge_output_format" is
339 replaced by this extension when given
340 fixup: Automatically correct known faults of the file.
342 - "never": do nothing
343 - "warn": only emit a warning
344 - "detect_or_warn": check whether we can do anything
345 about it, warn otherwise (default)
346 source_address: Client-side IP address to bind to.
347 call_home: Boolean, true iff we are allowed to contact the
348 yt-dlp servers for debugging. (BROKEN)
349 sleep_interval_requests: Number of seconds to sleep between requests
351 sleep_interval: Number of seconds to sleep before each download when
352 used alone or a lower bound of a range for randomized
353 sleep before each download (minimum possible number
354 of seconds to sleep) when used along with
356 max_sleep_interval:Upper bound of a range for randomized sleep before each
357 download (maximum possible number of seconds to sleep).
358 Must only be used along with sleep_interval.
359 Actual sleep time will be a random float from range
360 [sleep_interval; max_sleep_interval].
361 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
362 listformats: Print an overview of available video formats and exit.
363 list_thumbnails: Print a table of all thumbnails and exit.
364 match_filter: A function that gets called with the info_dict of
366 If it returns a message, the video is ignored.
367 If it returns None, the video is downloaded.
368 match_filter_func in utils.py is one example for this.
369 no_color: Do not emit color codes in output.
370 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
373 Two-letter ISO 3166-2 country code that will be used for
374 explicit geographic restriction bypassing via faking
375 X-Forwarded-For HTTP header
377 IP range in CIDR notation that will be used similarly to
380 The following options determine which downloader is picked:
381 external_downloader: A dictionary of protocol keys and the executable of the
382 external downloader to use for it. The allowed protocols
383 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
384 Set the value to 'native' to use the native downloader
385 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
386 or {'m3u8': 'ffmpeg'} instead.
387 Use the native HLS downloader instead of ffmpeg/avconv
388 if True, otherwise use ffmpeg/avconv if False, otherwise
389 use downloader suggested by extractor if None.
390 compat_opts: Compatibility options. See "Differences in default behavior".
391 Note that only format-sort, format-spec, no-live-chat,
392 no-attach-info-json, playlist-index, list-formats,
393 no-direct-merge, no-youtube-channel-redirect,
394 and no-youtube-unavailable-videos works when used via the API
396 The following parameters are not used by YoutubeDL itself, they are used by
397 the downloader (see yt_dlp/downloader/common.py):
398 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
399 noresizebuffer, retries, continuedl, noprogress, consoletitle,
400 xattr_set_filesize, external_downloader_args, hls_use_mpegts,
403 The following options are used by the post processors:
404 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
405 otherwise prefer ffmpeg. (avconv support is deprecated)
406 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
407 to the binary or its containing directory.
408 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
409 and a list of additional command-line arguments for the
410 postprocessor/executable. The dict can also have "PP+EXE" keys
411 which are used when the given exe is used by the given PP.
412 Use 'default' as the name for arguments to passed to all PP
414 The following options are used by the extractors:
415 extractor_retries: Number of times to retry for known errors
416 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
417 hls_split_discontinuity: Split HLS playlists to different formats at
418 discontinuities such as ad breaks (default: False)
419 youtube_include_dash_manifest: If True (default), DASH manifests and related
420 data will be downloaded and processed by extractor.
421 You can reduce network I/O by disabling it if you don't
422 care about DASH. (only for youtube)
423 youtube_include_hls_manifest: If True (default), HLS manifests and related
424 data will be downloaded and processed by extractor.
425 You can reduce network I/O by disabling it if you don't
426 care about HLS. (only for youtube)
429 _NUMERIC_FIELDS
= set((
430 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
431 'timestamp', 'upload_year', 'upload_month', 'upload_day',
432 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
433 'average_rating', 'comment_count', 'age_limit',
434 'start_time', 'end_time',
435 'chapter_number', 'season_number', 'episode_number',
436 'track_number', 'disc_number', 'release_year',
442 _pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
443 __prepare_filename_warned
= False
444 _first_webpage_request
= True
445 _download_retcode
= None
446 _num_downloads
= None
448 _playlist_urls
= set()
451 def __init__(self
, params
=None, auto_init
=True):
452 """Create a FileDownloader object with the given options."""
456 self
._ies
_instances
= {}
457 self
._pps
= {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
458 self
.__prepare
_filename
_warned
= False
459 self
._first
_webpage
_request
= True
460 self
._post
_hooks
= []
461 self
._progress
_hooks
= []
462 self
._download
_retcode
= 0
463 self
._num
_downloads
= 0
464 self
._screen
_file
= [sys
.stdout
, sys
.stderr
][params
.get('logtostderr', False)]
465 self
._err
_file
= sys
.stderr
468 'nocheckcertificate': False,
470 self
.params
.update(params
)
471 self
.cache
= Cache(self
)
473 if sys
.version_info
< (3, 6):
475 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys
.version_info
[:2])
477 def check_deprecated(param
, option
, suggestion
):
478 if self
.params
.get(param
) is not None:
479 self
.report_warning('%s is deprecated. Use %s instead' % (option
, suggestion
))
483 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
484 if self
.params
.get('geo_verification_proxy') is None:
485 self
.params
['geo_verification_proxy'] = self
.params
['cn_verification_proxy']
487 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
488 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
489 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
491 for msg
in self
.params
.get('warnings', []):
492 self
.report_warning(msg
)
494 if self
.params
.get('final_ext'):
495 if self
.params
.get('merge_output_format'):
496 self
.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
497 self
.params
['merge_output_format'] = self
.params
['final_ext']
499 if 'overwrites' in self
.params
and self
.params
['overwrites'] is None:
500 del self
.params
['overwrites']
502 if params
.get('bidi_workaround', False):
505 master
, slave
= pty
.openpty()
506 width
= compat_get_terminal_size().columns
510 width_args
= ['-w', str(width
)]
512 stdin
=subprocess
.PIPE
,
514 stderr
=self
._err
_file
)
516 self
._output
_process
= subprocess
.Popen(
517 ['bidiv'] + width_args
, **sp_kwargs
520 self
._output
_process
= subprocess
.Popen(
521 ['fribidi', '-c', 'UTF-8'] + width_args
, **sp_kwargs
)
522 self
._output
_channel
= os
.fdopen(master
, 'rb')
523 except OSError as ose
:
524 if ose
.errno
== errno
.ENOENT
:
525 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 $PATH.')
529 if (sys
.platform
!= 'win32'
530 and sys
.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
531 and not params
.get('restrictfilenames', False)):
532 # Unicode filesystem API will throw errors (#1474, #13027)
534 'Assuming --restrict-filenames since file system encoding '
535 'cannot encode all characters. '
536 'Set the LC_ALL environment variable to fix this.')
537 self
.params
['restrictfilenames'] = True
539 self
.outtmpl_dict
= self
.parse_outtmpl()
543 """Preload the archive, if any is specified"""
544 def preload_download_archive(fn
):
547 self
.write_debug('Loading archive file %r\n' % fn
)
549 with locked_file(fn
, 'r', encoding
='utf-8') as archive_file
:
550 for line
in archive_file
:
551 self
.archive
.add(line
.strip())
552 except IOError as ioe
:
553 if ioe
.errno
!= errno
.ENOENT
:
559 preload_download_archive(self
.params
.get('download_archive'))
562 self
.print_debug_header()
563 self
.add_default_info_extractors()
565 for pp_def_raw
in self
.params
.get('postprocessors', []):
566 pp_class
= get_postprocessor(pp_def_raw
['key'])
567 pp_def
= dict(pp_def_raw
)
570 when
= pp_def
['when']
573 when
= 'post_process'
574 pp
= pp_class(self
, **compat_kwargs(pp_def
))
575 self
.add_post_processor(pp
, when
=when
)
577 for ph
in self
.params
.get('post_hooks', []):
578 self
.add_post_hook(ph
)
580 for ph
in self
.params
.get('progress_hooks', []):
581 self
.add_progress_hook(ph
)
583 register_socks_protocols()
585 def warn_if_short_id(self
, argv
):
586 # short YouTube ID starting with dash?
588 i
for i
, a
in enumerate(argv
)
589 if re
.match(r
'^-[0-9A-Za-z_-]{10}$', a
)]
593 + [a
for i
, a
in enumerate(argv
) if i
not in idxs
]
594 + ['--'] + [argv
[i
] for i
in idxs
]
597 'Long argument string detected. '
598 'Use -- to separate parameters and URLs, like this:\n%s\n' %
599 args_to_str(correct_argv
))
601 def add_info_extractor(self
, ie
):
602 """Add an InfoExtractor object to the end of the list."""
604 if not isinstance(ie
, type):
605 self
._ies
_instances
[ie
.ie_key()] = ie
606 ie
.set_downloader(self
)
608 def get_info_extractor(self
, ie_key
):
610 Get an instance of an IE with name ie_key, it will try to get one from
611 the _ies list, if there's no instance it will create a new one and add
612 it to the extractor list.
614 ie
= self
._ies
_instances
.get(ie_key
)
616 ie
= get_info_extractor(ie_key
)()
617 self
.add_info_extractor(ie
)
620 def add_default_info_extractors(self
):
622 Add the InfoExtractors returned by gen_extractors to the end of the list
624 for ie
in gen_extractor_classes():
625 self
.add_info_extractor(ie
)
627 def add_post_processor(self
, pp
, when
='post_process'):
628 """Add a PostProcessor object to the end of the chain."""
629 self
._pps
[when
].append(pp
)
630 pp
.set_downloader(self
)
632 def add_post_hook(self
, ph
):
633 """Add the post hook"""
634 self
._post
_hooks
.append(ph
)
636 def add_progress_hook(self
, ph
):
637 """Add the progress hook (currently only for the file downloader)"""
638 self
._progress
_hooks
.append(ph
)
640 def _bidi_workaround(self
, message
):
641 if not hasattr(self
, '_output_channel'):
644 assert hasattr(self
, '_output_process')
645 assert isinstance(message
, compat_str
)
646 line_count
= message
.count('\n') + 1
647 self
._output
_process
.stdin
.write((message
+ '\n').encode('utf-8'))
648 self
._output
_process
.stdin
.flush()
649 res
= ''.join(self
._output
_channel
.readline().decode('utf-8')
650 for _
in range(line_count
))
651 return res
[:-len('\n')]
653 def _write_string(self
, s
, out
=None):
654 write_string(s
, out
=out
, encoding
=self
.params
.get('encoding'))
656 def to_stdout(self
, message
, skip_eol
=False, quiet
=False):
657 """Print message to stdout"""
658 if self
.params
.get('logger'):
659 self
.params
['logger'].debug(message
)
660 elif not quiet
or self
.params
.get('verbose'):
662 '%s%s' % (self
._bidi
_workaround
(message
), ('' if skip_eol
else '\n')),
663 self
._err
_file
if quiet
else self
._screen
_file
)
665 def to_stderr(self
, message
):
666 """Print message to stderr"""
667 assert isinstance(message
, compat_str
)
668 if self
.params
.get('logger'):
669 self
.params
['logger'].error(message
)
671 self
._write
_string
('%s\n' % self
._bidi
_workaround
(message
), self
._err
_file
)
673 def to_console_title(self
, message
):
674 if not self
.params
.get('consoletitle', False):
676 if compat_os_name
== 'nt':
677 if ctypes
.windll
.kernel32
.GetConsoleWindow():
678 # c_wchar_p() might not be necessary if `message` is
679 # already of type unicode()
680 ctypes
.windll
.kernel32
.SetConsoleTitleW(ctypes
.c_wchar_p(message
))
681 elif 'TERM' in os
.environ
:
682 self
._write
_string
('\033]0;%s\007' % message
, self
._screen
_file
)
684 def save_console_title(self
):
685 if not self
.params
.get('consoletitle', False):
687 if self
.params
.get('simulate', False):
689 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
690 # Save the title on stack
691 self
._write
_string
('\033[22;0t', self
._screen
_file
)
693 def restore_console_title(self
):
694 if not self
.params
.get('consoletitle', False):
696 if self
.params
.get('simulate', False):
698 if compat_os_name
!= 'nt' and 'TERM' in os
.environ
:
699 # Restore the title from stack
700 self
._write
_string
('\033[23;0t', self
._screen
_file
)
703 self
.save_console_title()
706 def __exit__(self
, *args
):
707 self
.restore_console_title()
709 if self
.params
.get('cookiefile') is not None:
710 self
.cookiejar
.save(ignore_discard
=True, ignore_expires
=True)
712 def trouble(self
, message
=None, tb
=None):
713 """Determine action to take when a download problem appears.
715 Depending on if the downloader has been configured to ignore
716 download errors or not, this method may throw an exception or
717 not when errors are found, after printing the message.
719 tb, if given, is additional traceback information.
721 if message
is not None:
722 self
.to_stderr(message
)
723 if self
.params
.get('verbose'):
725 if sys
.exc_info()[0]: # if .trouble has been called from an except block
727 if hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
728 tb
+= ''.join(traceback
.format_exception(*sys
.exc_info()[1].exc_info
))
729 tb
+= encode_compat_str(traceback
.format_exc())
731 tb_data
= traceback
.format_list(traceback
.extract_stack())
732 tb
= ''.join(tb_data
)
735 if not self
.params
.get('ignoreerrors', False):
736 if sys
.exc_info()[0] and hasattr(sys
.exc_info()[1], 'exc_info') and sys
.exc_info()[1].exc_info
[0]:
737 exc_info
= sys
.exc_info()[1].exc_info
739 exc_info
= sys
.exc_info()
740 raise DownloadError(message
, exc_info
)
741 self
._download
_retcode
= 1
743 def to_screen(self
, message
, skip_eol
=False):
744 """Print message to stdout if not in quiet mode"""
746 message
, skip_eol
, quiet
=self
.params
.get('quiet', False))
748 def report_warning(self
, message
):
750 Print the message to stderr, it will be prefixed with 'WARNING:'
751 If stderr is a tty file the 'WARNING:' will be colored
753 if self
.params
.get('logger') is not None:
754 self
.params
['logger'].warning(message
)
756 if self
.params
.get('no_warnings'):
758 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
759 _msg_header
= '\033[0;33mWARNING:\033[0m'
761 _msg_header
= 'WARNING:'
762 warning_message
= '%s %s' % (_msg_header
, message
)
763 self
.to_stderr(warning_message
)
765 def report_error(self
, message
, tb
=None):
767 Do the same as trouble, but prefixes the message with 'ERROR:', colored
768 in red if stderr is a tty file.
770 if not self
.params
.get('no_color') and self
._err
_file
.isatty() and compat_os_name
!= 'nt':
771 _msg_header
= '\033[0;31mERROR:\033[0m'
773 _msg_header
= 'ERROR:'
774 error_message
= '%s %s' % (_msg_header
, message
)
775 self
.trouble(error_message
, tb
)
777 def write_debug(self
, message
):
778 '''Log debug message or Print message to stderr'''
779 if not self
.params
.get('verbose', False):
781 message
= '[debug] %s' % message
782 if self
.params
.get('logger'):
783 self
.params
['logger'].debug(message
)
785 self
._write
_string
('%s\n' % message
)
787 def report_file_already_downloaded(self
, file_name
):
788 """Report file has already been fully downloaded."""
790 self
.to_screen('[download] %s has already been downloaded' % file_name
)
791 except UnicodeEncodeError:
792 self
.to_screen('[download] The file has already been downloaded')
794 def report_file_delete(self
, file_name
):
795 """Report that existing file will be deleted."""
797 self
.to_screen('Deleting existing file %s' % file_name
)
798 except UnicodeEncodeError:
799 self
.to_screen('Deleting existing file')
801 def parse_outtmpl(self
):
802 outtmpl_dict
= self
.params
.get('outtmpl', {})
803 if not isinstance(outtmpl_dict
, dict):
804 outtmpl_dict
= {'default': outtmpl_dict}
805 outtmpl_dict
.update({
806 k
: v
for k
, v
in DEFAULT_OUTTMPL
.items()
807 if not outtmpl_dict
.get(k
)})
808 for key
, val
in outtmpl_dict
.items():
809 if isinstance(val
, bytes):
811 'Parameter outtmpl is bytes, but should be a unicode string. '
812 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
816 def validate_outtmpl(tmpl
):
817 ''' @return None or Exception object '''
820 STR_FORMAT_RE
.format(''),
821 lambda mobj
: ('%' if not mobj
.group('has_key') else '') + mobj
.group(0),
823 ) % collections
.defaultdict(int)
825 except ValueError as err
:
828 def prepare_outtmpl(self
, outtmpl
, info_dict
, sanitize
=None):
829 """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)"""
830 info_dict
= dict(info_dict
)
831 na
= self
.params
.get('outtmpl_na_placeholder', 'NA')
833 info_dict
['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
834 formatSeconds(info_dict
['duration'], '-' if sanitize
else ':')
835 if info_dict
.get('duration', None) is not None
837 info_dict
['epoch'] = int(time
.time())
838 info_dict
['autonumber'] = self
.params
.get('autonumber_start', 1) - 1 + self
._num
_downloads
839 if info_dict
.get('resolution') is None:
840 info_dict
['resolution'] = self
.format_resolution(info_dict
, default
=None)
842 # For fields playlist_index and autonumber convert all occurrences
843 # of %(field)s to %(field)0Nd for backward compatibility
844 field_size_compat_map
= {
845 'playlist_index': len(str(info_dict
.get('_last_playlist_index') or '')),
846 'autonumber': self
.params
.get('autonumber_size') or 5,
849 EXTERNAL_FORMAT_RE
= STR_FORMAT_RE
.format('[^)]*')
850 # Field is of the form key1.key2...
851 # where keys (except first) can be string, int or slice
852 FIELD_RE
= r
'\w+(?:\.(?:\w+|[-\d]*(?::[-\d]*){0,2}))*'
853 INTERNAL_FORMAT_RE
= re
.compile(r
'''(?x)
856 (?P<maths>(?:[-+]-?(?:\d+(?:\.\d+)?|{0}))*)
857 (?:>(?P<strf_format>.+?))?
858 (?:\|(?P<default>.*?))?
859 $'''.format(FIELD_RE
))
860 MATH_OPERATORS_RE
= re
.compile(r
'(?<![-+])([-+])')
867 get_key
= lambda k
: traverse_obj(
868 info_dict
, k
.split('.'), is_user_input
=True, traverse_string
=True)
870 def get_value(mdict
):
872 value
= get_key(mdict
['fields'])
875 value
= float_or_none(value
)
876 if value
is not None:
880 value
= float_or_none(value
)
882 for item
in MATH_OPERATORS_RE
.split(mdict
['maths'])[1:]:
883 if item
== '' or value
is None:
886 item
, multiplier
= (item
[1:], -1) if item
[0] == '-' else (item
, 1)
887 offset
= float_or_none(item
)
889 offset
= float_or_none(get_key(item
))
891 value
= operator(value
, multiplier
* offset
)
892 except (TypeError, ZeroDivisionError):
896 operator
= MATH_FUNCTIONS
[item
]
897 # Datetime formatting
898 if mdict
['strf_format']:
899 value
= strftime_or_none(value
, mdict
['strf_format'])
903 def create_key(outer_mobj
):
904 if not outer_mobj
.group('has_key'):
905 return '%{}'.format(outer_mobj
.group(0))
907 key
= outer_mobj
.group('key')
908 fmt
= outer_mobj
.group('format')
909 mobj
= re
.match(INTERNAL_FORMAT_RE
, key
)
911 value
, default
= None, na
913 mobj
= mobj
.groupdict()
914 default
= mobj
['default'] if mobj
['default'] is not None else na
915 value
= get_value(mobj
)
917 if fmt
== 's' and value
is not None and key
in field_size_compat_map
.keys():
918 fmt
= '0{:d}d'.format(field_size_compat_map
[key
])
920 value
= default
if value
is None else value
924 value
= compat_str(value
)
926 value
, fmt
= default
, 's'
929 elif fmt
[-1] not in 'rs': # numeric
930 value
= float_or_none(value
)
932 value
, fmt
= default
, 's'
935 # If value is an object, sanitize might convert it to a string
936 # So we convert it to repr first
937 value
, fmt
= repr(value
), '%ss' % fmt
[:-1]
938 value
= sanitize(key
, value
)
939 tmpl_dict
[key
] = value
940 return '%({key}){fmt}'.format(key
=key
, fmt
=fmt
)
942 return re
.sub(EXTERNAL_FORMAT_RE
, create_key
, outtmpl
), tmpl_dict
944 def _prepare_filename(self
, info_dict
, tmpl_type
='default'):
946 sanitize
= lambda k
, v
: sanitize_filename(
948 restricted
=self
.params
.get('restrictfilenames'),
949 is_id
=(k
== 'id' or k
.endswith('_id')))
950 outtmpl
= self
.outtmpl_dict
.get(tmpl_type
, self
.outtmpl_dict
['default'])
951 outtmpl
, template_dict
= self
.prepare_outtmpl(outtmpl
, info_dict
, sanitize
)
953 # expand_path translates '%%' into '%' and '$$' into '$'
954 # correspondingly that is not what we want since we need to keep
955 # '%%' intact for template dict substitution step. Working around
956 # with boundary-alike separator hack.
957 sep
= ''.join([random
.choice(ascii_letters
) for _
in range(32)])
958 outtmpl
= outtmpl
.replace('%%', '%{0}%'.format(sep
)).replace('$$', '${0}$'.format(sep
))
960 # outtmpl should be expand_path'ed before template dict substitution
961 # because meta fields may contain env variables we don't want to
962 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
963 # title "Hello $PATH", we don't want `$PATH` to be expanded.
964 filename
= expand_path(outtmpl
).replace(sep
, '') % template_dict
966 force_ext
= OUTTMPL_TYPES
.get(tmpl_type
)
967 if force_ext
is not None:
968 filename
= replace_extension(filename
, force_ext
, info_dict
.get('ext'))
970 # https://github.com/blackjack4494/youtube-dlc/issues/85
971 trim_file_name
= self
.params
.get('trim_file_name', False)
973 fn_groups
= filename
.rsplit('.')
976 if len(fn_groups
) > 2:
977 sub_ext
= fn_groups
[-2]
978 filename
= '.'.join(filter(None, [fn_groups
[0][:trim_file_name
], sub_ext
, ext
]))
981 except ValueError as err
:
982 self
.report_error('Error in output template: ' + str(err
) + ' (encoding: ' + repr(preferredencoding()) + ')')
985 def prepare_filename(self
, info_dict
, dir_type
='', warn
=False):
986 """Generate the output filename."""
987 paths
= self
.params
.get('paths', {})
988 assert isinstance(paths
, dict)
989 filename
= self
._prepare
_filename
(info_dict
, dir_type
or 'default')
991 if warn
and not self
.__prepare
_filename
_warned
:
994 elif filename
== '-':
995 self
.report_warning('--paths is ignored when an outputting to stdout')
996 elif os
.path
.isabs(filename
):
997 self
.report_warning('--paths is ignored since an absolute path is given in output template')
998 self
.__prepare
_filename
_warned
= True
999 if filename
== '-' or not filename
:
1002 homepath
= expand_path(paths
.get('home', '').strip())
1003 assert isinstance(homepath
, compat_str
)
1004 subdir
= expand_path(paths
.get(dir_type
, '').strip()) if dir_type
else ''
1005 assert isinstance(subdir
, compat_str
)
1006 path
= os
.path
.join(homepath
, subdir
, filename
)
1008 # Temporary fix for #4787
1009 # 'Treat' all problem characters by passing filename through preferredencoding
1010 # to workaround encoding issues with subprocess on python2 @ Windows
1011 if sys
.version_info
< (3, 0) and sys
.platform
== 'win32':
1012 path
= encodeFilename(path
, True).decode(preferredencoding())
1013 return sanitize_path(path
, force
=self
.params
.get('windowsfilenames'))
1015 def _match_entry(self
, info_dict
, incomplete
=False, silent
=False):
1016 """ Returns None if the file should be downloaded """
1018 video_title
= info_dict
.get('title', info_dict
.get('id', 'video'))
1021 if 'title' in info_dict
:
1022 # This can happen when we're just evaluating the playlist
1023 title
= info_dict
['title']
1024 matchtitle
= self
.params
.get('matchtitle', False)
1026 if not re
.search(matchtitle
, title
, re
.IGNORECASE
):
1027 return '"' + title
+ '" title did not match pattern "' + matchtitle
+ '"'
1028 rejecttitle
= self
.params
.get('rejecttitle', False)
1030 if re
.search(rejecttitle
, title
, re
.IGNORECASE
):
1031 return '"' + title
+ '" title matched reject pattern "' + rejecttitle
+ '"'
1032 date
= info_dict
.get('upload_date')
1033 if date
is not None:
1034 dateRange
= self
.params
.get('daterange', DateRange())
1035 if date
not in dateRange
:
1036 return '%s upload date is not in range %s' % (date_from_str(date
).isoformat(), dateRange
)
1037 view_count
= info_dict
.get('view_count')
1038 if view_count
is not None:
1039 min_views
= self
.params
.get('min_views')
1040 if min_views
is not None and view_count
< min_views
:
1041 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title
, view_count
, min_views
)
1042 max_views
= self
.params
.get('max_views')
1043 if max_views
is not None and view_count
> max_views
:
1044 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title
, view_count
, max_views
)
1045 if age_restricted(info_dict
.get('age_limit'), self
.params
.get('age_limit')):
1046 return 'Skipping "%s" because it is age restricted' % video_title
1049 match_filter
= self
.params
.get('match_filter')
1050 if match_filter
is not None:
1051 ret
= match_filter(info_dict
)
1056 if self
.in_download_archive(info_dict
):
1057 reason
= '%s has already been recorded in the archive' % video_title
1058 break_opt
, break_err
= 'break_on_existing', ExistingVideoReached
1060 reason
= check_filter()
1061 break_opt
, break_err
= 'break_on_reject', RejectedVideoReached
1062 if reason
is not None:
1064 self
.to_screen('[download] ' + reason
)
1065 if self
.params
.get(break_opt
, False):
1070 def add_extra_info(info_dict
, extra_info
):
1071 '''Set the keys from extra_info in info dict if they are missing'''
1072 for key
, value
in extra_info
.items():
1073 info_dict
.setdefault(key
, value
)
1075 def extract_info(self
, url
, download
=True, ie_key
=None, extra_info
={},
1076 process
=True, force_generic_extractor
=False):
1078 Return a list with a dictionary for each video extracted.
1081 url -- URL to extract
1084 download -- whether to download videos during extraction
1085 ie_key -- extractor key hint
1086 extra_info -- dictionary containing the extra values to add to each result
1087 process -- whether to resolve all unresolved references (URLs, playlist items),
1088 must be True for download to work.
1089 force_generic_extractor -- force using the generic extractor
1092 if not ie_key
and force_generic_extractor
:
1096 ies
= [self
.get_info_extractor(ie_key
)]
1101 if not ie
.suitable(url
):
1104 ie_key
= ie
.ie_key()
1105 ie
= self
.get_info_extractor(ie_key
)
1106 if not ie
.working():
1107 self
.report_warning('The program functionality for this site has been marked as broken, '
1108 'and will probably not work.')
1111 temp_id
= str_or_none(
1112 ie
.extract_id(url
) if callable(getattr(ie
, 'extract_id', None))
1113 else ie
._match
_id
(url
))
1114 except (AssertionError, IndexError, AttributeError):
1116 if temp_id
is not None and self
.in_download_archive({'id': temp_id, 'ie_key': ie_key}
):
1117 self
.to_screen("[%s] %s: has already been recorded in archive" % (
1120 return self
.__extract
_info
(url
, ie
, download
, extra_info
, process
)
1122 self
.report_error('no suitable InfoExtractor for URL %s' % url
)
1124 def __handle_extraction_exceptions(func
):
1125 def wrapper(self
, *args
, **kwargs
):
1127 return func(self
, *args
, **kwargs
)
1128 except GeoRestrictedError
as e
:
1131 msg
+= '\nThis video is available in %s.' % ', '.join(
1132 map(ISO3166Utils
.short2full
, e
.countries
))
1133 msg
+= '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1134 self
.report_error(msg
)
1135 except ExtractorError
as e
: # An error we somewhat expected
1136 self
.report_error(compat_str(e
), e
.format_traceback())
1137 except (MaxDownloadsReached
, ExistingVideoReached
, RejectedVideoReached
):
1139 except Exception as e
:
1140 if self
.params
.get('ignoreerrors', False):
1141 self
.report_error(error_to_compat_str(e
), tb
=encode_compat_str(traceback
.format_exc()))
1146 @__handle_extraction_exceptions
1147 def __extract_info(self
, url
, ie
, download
, extra_info
, process
):
1148 ie_result
= ie
.extract(url
)
1149 if ie_result
is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1151 if isinstance(ie_result
, list):
1152 # Backwards compatibility: old IE result format
1154 '_type': 'compat_list',
1155 'entries': ie_result
,
1157 self
.add_default_extra_info(ie_result
, ie
, url
)
1159 return self
.process_ie_result(ie_result
, download
, extra_info
)
1163 def add_default_extra_info(self
, ie_result
, ie
, url
):
1164 self
.add_extra_info(ie_result
, {
1165 'extractor': ie
.IE_NAME
,
1167 'original_url': url
,
1168 'webpage_url_basename': url_basename(url
),
1169 'extractor_key': ie
.ie_key(),
1172 def process_ie_result(self
, ie_result
, download
=True, extra_info
={}):
1174 Take the result of the ie(may be modified) and resolve all unresolved
1175 references (URLs, playlist items).
1177 It will also download the videos if 'download'.
1178 Returns the resolved ie_result.
1180 result_type
= ie_result
.get('_type', 'video')
1182 if result_type
in ('url', 'url_transparent'):
1183 ie_result
['url'] = sanitize_url(ie_result
['url'])
1184 extract_flat
= self
.params
.get('extract_flat', False)
1185 if ((extract_flat
== 'in_playlist' and 'playlist' in extra_info
)
1186 or extract_flat
is True):
1187 info_copy
= ie_result
.copy()
1188 self
.add_extra_info(info_copy
, extra_info
)
1189 self
.add_default_extra_info(
1190 info_copy
, self
.get_info_extractor(ie_result
.get('ie_key')), ie_result
['url'])
1191 self
.__forced
_printings
(info_copy
, self
.prepare_filename(info_copy
), incomplete
=True)
1194 if result_type
== 'video':
1195 self
.add_extra_info(ie_result
, extra_info
)
1196 ie_result
= self
.process_video_result(ie_result
, download
=download
)
1197 additional_urls
= (ie_result
or {}).get('additional_urls')
1199 # TODO: Improve MetadataFromFieldPP to allow setting a list
1200 if isinstance(additional_urls
, compat_str
):
1201 additional_urls
= [additional_urls
]
1203 '[info] %s: %d additional URL(s) requested' % (ie_result
['id'], len(additional_urls
)))
1204 self
.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls
))
1205 ie_result
['additional_entries'] = [
1207 url
, download
, extra_info
,
1208 force_generic_extractor
=self
.params
.get('force_generic_extractor'))
1209 for url
in additional_urls
1212 elif result_type
== 'url':
1213 # We have to add extra_info to the results because it may be
1214 # contained in a playlist
1215 return self
.extract_info(
1216 ie_result
['url'], download
,
1217 ie_key
=ie_result
.get('ie_key'),
1218 extra_info
=extra_info
)
1219 elif result_type
== 'url_transparent':
1220 # Use the information from the embedding page
1221 info
= self
.extract_info(
1222 ie_result
['url'], ie_key
=ie_result
.get('ie_key'),
1223 extra_info
=extra_info
, download
=False, process
=False)
1225 # extract_info may return None when ignoreerrors is enabled and
1226 # extraction failed with an error, don't crash and return early
1231 force_properties
= dict(
1232 (k
, v
) for k
, v
in ie_result
.items() if v
is not None)
1233 for f
in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1234 if f
in force_properties
:
1235 del force_properties
[f
]
1236 new_result
= info
.copy()
1237 new_result
.update(force_properties
)
1239 # Extracted info may not be a video result (i.e.
1240 # info.get('_type', 'video') != video) but rather an url or
1241 # url_transparent. In such cases outer metadata (from ie_result)
1242 # should be propagated to inner one (info). For this to happen
1243 # _type of info should be overridden with url_transparent. This
1244 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1245 if new_result
.get('_type') == 'url':
1246 new_result
['_type'] = 'url_transparent'
1248 return self
.process_ie_result(
1249 new_result
, download
=download
, extra_info
=extra_info
)
1250 elif result_type
in ('playlist', 'multi_video'):
1251 # Protect from infinite recursion due to recursively nested playlists
1252 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1253 webpage_url
= ie_result
['webpage_url']
1254 if webpage_url
in self
._playlist
_urls
:
1256 '[download] Skipping already downloaded playlist: %s'
1257 % ie_result
.get('title') or ie_result
.get('id'))
1260 self
._playlist
_level
+= 1
1261 self
._playlist
_urls
.add(webpage_url
)
1262 self
._sanitize
_thumbnails
(ie_result
)
1264 return self
.__process
_playlist
(ie_result
, download
)
1266 self
._playlist
_level
-= 1
1267 if not self
._playlist
_level
:
1268 self
._playlist
_urls
.clear()
1269 elif result_type
== 'compat_list':
1270 self
.report_warning(
1271 'Extractor %s returned a compat_list result. '
1272 'It needs to be updated.' % ie_result
.get('extractor'))
1275 self
.add_extra_info(
1278 'extractor': ie_result
['extractor'],
1279 'webpage_url': ie_result
['webpage_url'],
1280 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1281 'extractor_key': ie_result
['extractor_key'],
1285 ie_result
['entries'] = [
1286 self
.process_ie_result(_fixup(r
), download
, extra_info
)
1287 for r
in ie_result
['entries']
1291 raise Exception('Invalid result type: %s' % result_type
)
1293 def _ensure_dir_exists(self
, path
):
1294 return make_dir(path
, self
.report_error
)
1296 def __process_playlist(self
, ie_result
, download
):
1297 # We process each entry in the playlist
1298 playlist
= ie_result
.get('title') or ie_result
.get('id')
1299 self
.to_screen('[download] Downloading playlist: %s' % playlist
)
1301 if 'entries' not in ie_result
:
1302 raise EntryNotInPlaylist()
1303 incomplete_entries
= bool(ie_result
.get('requested_entries'))
1304 if incomplete_entries
:
1305 def fill_missing_entries(entries
, indexes
):
1306 ret
= [None] * max(*indexes
)
1307 for i
, entry
in zip(indexes
, entries
):
1310 ie_result
['entries'] = fill_missing_entries(ie_result
['entries'], ie_result
['requested_entries'])
1312 playlist_results
= []
1314 playliststart
= self
.params
.get('playliststart', 1)
1315 playlistend
= self
.params
.get('playlistend')
1316 # For backwards compatibility, interpret -1 as whole list
1317 if playlistend
== -1:
1320 playlistitems_str
= self
.params
.get('playlist_items')
1321 playlistitems
= None
1322 if playlistitems_str
is not None:
1323 def iter_playlistitems(format
):
1324 for string_segment
in format
.split(','):
1325 if '-' in string_segment
:
1326 start
, end
= string_segment
.split('-')
1327 for item
in range(int(start
), int(end
) + 1):
1330 yield int(string_segment
)
1331 playlistitems
= orderedSet(iter_playlistitems(playlistitems_str
))
1333 ie_entries
= ie_result
['entries']
1335 'Downloading %d videos' if not isinstance(ie_entries
, list)
1336 else 'Collected %d videos; downloading %%d of them' % len(ie_entries
))
1337 if not isinstance(ie_entries
, (list, PagedList
)):
1338 ie_entries
= LazyList(ie_entries
)
1341 for i
in playlistitems
or itertools
.count(playliststart
):
1342 if playlistitems
is None and playlistend
is not None and playlistend
< i
:
1346 entry
= ie_entries
[i
- 1]
1348 raise EntryNotInPlaylist()
1349 except (IndexError, EntryNotInPlaylist
):
1350 if incomplete_entries
:
1351 raise EntryNotInPlaylist()
1352 elif not playlistitems
:
1354 entries
.append(entry
)
1356 if entry
is not None:
1357 self
._match
_entry
(entry
, incomplete
=True, silent
=True)
1358 except (ExistingVideoReached
, RejectedVideoReached
):
1360 ie_result
['entries'] = entries
1362 # Save playlist_index before re-ordering
1364 ((playlistitems
[i
- 1] if playlistitems
else i
), entry
)
1365 for i
, entry
in enumerate(entries
, 1)
1366 if entry
is not None]
1367 n_entries
= len(entries
)
1369 if not playlistitems
and (playliststart
or playlistend
):
1370 playlistitems
= list(range(playliststart
, playliststart
+ n_entries
))
1371 ie_result
['requested_entries'] = playlistitems
1373 if self
.params
.get('allow_playlist_files', True):
1375 'playlist': playlist
,
1376 'playlist_id': ie_result
.get('id'),
1377 'playlist_title': ie_result
.get('title'),
1378 'playlist_uploader': ie_result
.get('uploader'),
1379 'playlist_uploader_id': ie_result
.get('uploader_id'),
1380 'playlist_index': 0,
1382 ie_copy
.update(dict(ie_result
))
1384 if self
.params
.get('writeinfojson', False):
1385 infofn
= self
.prepare_filename(ie_copy
, 'pl_infojson')
1386 if not self
._ensure
_dir
_exists
(encodeFilename(infofn
)):
1388 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(infofn
)):
1389 self
.to_screen('[info] Playlist metadata is already present')
1391 self
.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn
)
1393 write_json_file(self
.filter_requested_info(ie_result
, self
.params
.get('clean_infojson', True)), infofn
)
1394 except (OSError, IOError):
1395 self
.report_error('Cannot write playlist metadata to JSON file ' + infofn
)
1397 # TODO: This should be passed to ThumbnailsConvertor if necessary
1398 self
._write
_thumbnails
(ie_copy
, self
.prepare_filename(ie_copy
, 'pl_thumbnail'))
1400 if self
.params
.get('writedescription', False):
1401 descfn
= self
.prepare_filename(ie_copy
, 'pl_description')
1402 if not self
._ensure
_dir
_exists
(encodeFilename(descfn
)):
1404 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(descfn
)):
1405 self
.to_screen('[info] Playlist description is already present')
1406 elif ie_result
.get('description') is None:
1407 self
.report_warning('There\'s no playlist description to write.')
1410 self
.to_screen('[info] Writing playlist description to: ' + descfn
)
1411 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
1412 descfile
.write(ie_result
['description'])
1413 except (OSError, IOError):
1414 self
.report_error('Cannot write playlist description file ' + descfn
)
1417 if self
.params
.get('playlistreverse', False):
1418 entries
= entries
[::-1]
1419 if self
.params
.get('playlistrandom', False):
1420 random
.shuffle(entries
)
1422 x_forwarded_for
= ie_result
.get('__x_forwarded_for_ip')
1424 self
.to_screen('[%s] playlist %s: %s' % (ie_result
['extractor'], playlist
, msg
% n_entries
))
1426 max_failures
= self
.params
.get('skip_playlist_after_errors') or float('inf')
1427 for i
, entry_tuple
in enumerate(entries
, 1):
1428 playlist_index
, entry
= entry_tuple
1429 if 'playlist_index' in self
.params
.get('compat_options', []):
1430 playlist_index
= playlistitems
[i
- 1] if playlistitems
else i
1431 self
.to_screen('[download] Downloading video %s of %s' % (i
, n_entries
))
1432 # This __x_forwarded_for_ip thing is a bit ugly but requires
1435 entry
['__x_forwarded_for_ip'] = x_forwarded_for
1437 'n_entries': n_entries
,
1438 '_last_playlist_index': max(playlistitems
) if playlistitems
else (playlistend
or n_entries
),
1439 'playlist_index': playlist_index
,
1440 'playlist_autonumber': i
,
1441 'playlist': playlist
,
1442 'playlist_id': ie_result
.get('id'),
1443 'playlist_title': ie_result
.get('title'),
1444 'playlist_uploader': ie_result
.get('uploader'),
1445 'playlist_uploader_id': ie_result
.get('uploader_id'),
1446 'extractor': ie_result
['extractor'],
1447 'webpage_url': ie_result
['webpage_url'],
1448 'webpage_url_basename': url_basename(ie_result
['webpage_url']),
1449 'extractor_key': ie_result
['extractor_key'],
1452 if self
._match
_entry
(entry
, incomplete
=True) is not None:
1455 entry_result
= self
.__process
_iterable
_entry
(entry
, download
, extra
)
1456 if not entry_result
:
1458 if failures
>= max_failures
:
1460 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist
, failures
))
1462 # TODO: skip failed (empty) entries?
1463 playlist_results
.append(entry_result
)
1464 ie_result
['entries'] = playlist_results
1465 self
.to_screen('[download] Finished downloading playlist: %s' % playlist
)
1468 @__handle_extraction_exceptions
1469 def __process_iterable_entry(self
, entry
, download
, extra_info
):
1470 return self
.process_ie_result(
1471 entry
, download
=download
, extra_info
=extra_info
)
1473 def _build_format_filter(self
, filter_spec
):
1474 " Returns a function to filter the formats according to the filter_spec "
1484 operator_rex
= re
.compile(r
'''(?x)\s*
1485 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
1486 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1487 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
1489 ''' % '|'.join(map(re
.escape
, OPERATORS
.keys())))
1490 m
= operator_rex
.search(filter_spec
)
1493 comparison_value
= int(m
.group('value'))
1495 comparison_value
= parse_filesize(m
.group('value'))
1496 if comparison_value
is None:
1497 comparison_value
= parse_filesize(m
.group('value') + 'B')
1498 if comparison_value
is None:
1500 'Invalid value %r in format specification %r' % (
1501 m
.group('value'), filter_spec
))
1502 op
= OPERATORS
[m
.group('op')]
1507 '^=': lambda attr
, value
: attr
.startswith(value
),
1508 '$=': lambda attr
, value
: attr
.endswith(value
),
1509 '*=': lambda attr
, value
: value
in attr
,
1511 str_operator_rex
= re
.compile(r
'''(?x)
1512 \s*(?P<key>[a-zA-Z0-9._-]+)
1513 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
1514 \s*(?P<value>[a-zA-Z0-9._-]+)
1516 ''' % '|'.join(map(re
.escape
, STR_OPERATORS
.keys())))
1517 m
= str_operator_rex
.search(filter_spec
)
1519 comparison_value
= m
.group('value')
1520 str_op
= STR_OPERATORS
[m
.group('op')]
1521 if m
.group('negation'):
1522 op
= lambda attr
, value
: not str_op(attr
, value
)
1527 raise ValueError('Invalid filter specification %r' % filter_spec
)
1530 actual_value
= f
.get(m
.group('key'))
1531 if actual_value
is None:
1532 return m
.group('none_inclusive')
1533 return op(actual_value
, comparison_value
)
1536 def _default_format_spec(self
, info_dict
, download
=True):
1539 merger
= FFmpegMergerPP(self
)
1540 return merger
.available
and merger
.can_merge()
1543 not self
.params
.get('simulate', False)
1547 or info_dict
.get('is_live', False)
1548 or self
.outtmpl_dict
['default'] == '-'))
1551 or self
.params
.get('allow_multiple_audio_streams', False)
1552 or 'format-spec' in self
.params
.get('compat_opts', []))
1555 'best/bestvideo+bestaudio' if prefer_best
1556 else 'bestvideo*+bestaudio/best' if not compat
1557 else 'bestvideo+bestaudio/best')
1559 def build_format_selector(self
, format_spec
):
1560 def syntax_error(note
, start
):
1562 'Invalid format specification: '
1563 '{0}\n\t{1}\n\t{2}^'.format(note
, format_spec
, ' ' * start
[1]))
1564 return SyntaxError(message
)
1566 PICKFIRST
= 'PICKFIRST'
1570 FormatSelector
= collections
.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1572 allow_multiple_streams
= {'audio': self
.params
.get('allow_multiple_audio_streams', False),
1573 'video': self
.params
.get('allow_multiple_video_streams', False)}
1575 check_formats
= self
.params
.get('check_formats')
1577 def _parse_filter(tokens
):
1579 for type, string
, start
, _
, _
in tokens
:
1580 if type == tokenize
.OP
and string
== ']':
1581 return ''.join(filter_parts
)
1583 filter_parts
.append(string
)
1585 def _remove_unused_ops(tokens
):
1586 # Remove operators that we don't use and join them with the surrounding strings
1587 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1588 ALLOWED_OPS
= ('/', '+', ',', '(', ')')
1589 last_string
, last_start
, last_end
, last_line
= None, None, None, None
1590 for type, string
, start
, end
, line
in tokens
:
1591 if type == tokenize
.OP
and string
== '[':
1593 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1595 yield type, string
, start
, end
, line
1596 # everything inside brackets will be handled by _parse_filter
1597 for type, string
, start
, end
, line
in tokens
:
1598 yield type, string
, start
, end
, line
1599 if type == tokenize
.OP
and string
== ']':
1601 elif type == tokenize
.OP
and string
in ALLOWED_OPS
:
1603 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1605 yield type, string
, start
, end
, line
1606 elif type in [tokenize
.NAME
, tokenize
.NUMBER
, tokenize
.OP
]:
1608 last_string
= string
1612 last_string
+= string
1614 yield tokenize
.NAME
, last_string
, last_start
, last_end
, last_line
1616 def _parse_format_selection(tokens
, inside_merge
=False, inside_choice
=False, inside_group
=False):
1618 current_selector
= None
1619 for type, string
, start
, _
, _
in tokens
:
1620 # ENCODING is only defined in python 3.x
1621 if type == getattr(tokenize
, 'ENCODING', None):
1623 elif type in [tokenize
.NAME
, tokenize
.NUMBER
]:
1624 current_selector
= FormatSelector(SINGLE
, string
, [])
1625 elif type == tokenize
.OP
:
1627 if not inside_group
:
1628 # ')' will be handled by the parentheses group
1629 tokens
.restore_last_token()
1631 elif inside_merge
and string
in ['/', ',']:
1632 tokens
.restore_last_token()
1634 elif inside_choice
and string
== ',':
1635 tokens
.restore_last_token()
1638 if not current_selector
:
1639 raise syntax_error('"," must follow a format selector', start
)
1640 selectors
.append(current_selector
)
1641 current_selector
= None
1643 if not current_selector
:
1644 raise syntax_error('"/" must follow a format selector', start
)
1645 first_choice
= current_selector
1646 second_choice
= _parse_format_selection(tokens
, inside_choice
=True)
1647 current_selector
= FormatSelector(PICKFIRST
, (first_choice
, second_choice
), [])
1649 if not current_selector
:
1650 current_selector
= FormatSelector(SINGLE
, 'best', [])
1651 format_filter
= _parse_filter(tokens
)
1652 current_selector
.filters
.append(format_filter
)
1654 if current_selector
:
1655 raise syntax_error('Unexpected "("', start
)
1656 group
= _parse_format_selection(tokens
, inside_group
=True)
1657 current_selector
= FormatSelector(GROUP
, group
, [])
1659 if not current_selector
:
1660 raise syntax_error('Unexpected "+"', start
)
1661 selector_1
= current_selector
1662 selector_2
= _parse_format_selection(tokens
, inside_merge
=True)
1664 raise syntax_error('Expected a selector', start
)
1665 current_selector
= FormatSelector(MERGE
, (selector_1
, selector_2
), [])
1667 raise syntax_error('Operator not recognized: "{0}"'.format(string
), start
)
1668 elif type == tokenize
.ENDMARKER
:
1670 if current_selector
:
1671 selectors
.append(current_selector
)
1674 def _merge(formats_pair
):
1675 format_1
, format_2
= formats_pair
1678 formats_info
.extend(format_1
.get('requested_formats', (format_1
,)))
1679 formats_info
.extend(format_2
.get('requested_formats', (format_2
,)))
1681 if not allow_multiple_streams
['video'] or not allow_multiple_streams
['audio']:
1682 get_no_more
= {"video": False, "audio": False}
1683 for (i
, fmt_info
) in enumerate(formats_info
):
1684 for aud_vid
in ["audio", "video"]:
1685 if not allow_multiple_streams
[aud_vid
] and fmt_info
.get(aud_vid
[0] + 'codec') != 'none':
1686 if get_no_more
[aud_vid
]:
1688 get_no_more
[aud_vid
] = True
1690 if len(formats_info
) == 1:
1691 return formats_info
[0]
1693 video_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('vcodec') != 'none']
1694 audio_fmts
= [fmt_info
for fmt_info
in formats_info
if fmt_info
.get('acodec') != 'none']
1696 the_only_video
= video_fmts
[0] if len(video_fmts
) == 1 else None
1697 the_only_audio
= audio_fmts
[0] if len(audio_fmts
) == 1 else None
1699 output_ext
= self
.params
.get('merge_output_format')
1702 output_ext
= the_only_video
['ext']
1703 elif the_only_audio
and not video_fmts
:
1704 output_ext
= the_only_audio
['ext']
1709 'requested_formats': formats_info
,
1710 'format': '+'.join(fmt_info
.get('format') for fmt_info
in formats_info
),
1711 'format_id': '+'.join(fmt_info
.get('format_id') for fmt_info
in formats_info
),
1717 'width': the_only_video
.get('width'),
1718 'height': the_only_video
.get('height'),
1719 'resolution': the_only_video
.get('resolution') or self
.format_resolution(the_only_video
),
1720 'fps': the_only_video
.get('fps'),
1721 'vcodec': the_only_video
.get('vcodec'),
1722 'vbr': the_only_video
.get('vbr'),
1723 'stretched_ratio': the_only_video
.get('stretched_ratio'),
1728 'acodec': the_only_audio
.get('acodec'),
1729 'abr': the_only_audio
.get('abr'),
1734 def _check_formats(formats
):
1736 self
.to_screen('[info] Testing format %s' % f
['format_id'])
1737 paths
= self
.params
.get('paths', {})
1738 temp_file
= os
.path
.join(
1739 expand_path(paths
.get('home', '').strip()),
1740 expand_path(paths
.get('temp', '').strip()),
1741 'ytdl.%s.f%s.check-format' % (random_uuidv4(), f
['format_id']))
1743 dl
, _
= self
.dl(temp_file
, f
, test
=True)
1744 except (ExtractorError
, IOError, OSError, ValueError) + network_exceptions
:
1747 if os
.path
.exists(temp_file
):
1748 os
.remove(temp_file
)
1752 self
.to_screen('[info] Unable to download format %s. Skipping...' % f
['format_id'])
1754 def _build_selector_function(selector
):
1755 if isinstance(selector
, list): # ,
1756 fs
= [_build_selector_function(s
) for s
in selector
]
1758 def selector_function(ctx
):
1760 for format
in f(ctx
):
1762 return selector_function
1764 elif selector
.type == GROUP
: # ()
1765 selector_function
= _build_selector_function(selector
.selector
)
1767 elif selector
.type == PICKFIRST
: # /
1768 fs
= [_build_selector_function(s
) for s
in selector
.selector
]
1770 def selector_function(ctx
):
1772 picked_formats
= list(f(ctx
))
1774 return picked_formats
1777 elif selector
.type == SINGLE
: # atom
1778 format_spec
= selector
.selector
or 'best'
1780 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
1781 if format_spec
== 'all':
1782 def selector_function(ctx
):
1783 formats
= list(ctx
['formats'])
1785 formats
= _check_formats(formats
)
1788 elif format_spec
== 'mergeall':
1789 def selector_function(ctx
):
1790 formats
= list(_check_formats(ctx
['formats']))
1793 merged_format
= formats
[-1]
1794 for f
in formats
[-2::-1]:
1795 merged_format
= _merge((merged_format
, f
))
1799 format_fallback
, format_reverse
, format_idx
= False, True, 1
1801 r
'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1803 if mobj
is not None:
1804 format_idx
= int_or_none(mobj
.group('n'), default
=1)
1805 format_reverse
= mobj
.group('bw')[0] == 'b'
1806 format_type
= (mobj
.group('type') or [None])[0]
1807 not_format_type
= {'v': 'a', 'a': 'v'}
.get(format_type
)
1808 format_modified
= mobj
.group('mod') is not None
1810 format_fallback
= not format_type
and not format_modified
# for b, w
1812 (lambda f
: f
.get('%scodec' % format_type
) != 'none')
1813 if format_type
and format_modified
# bv*, ba*, wv*, wa*
1814 else (lambda f
: f
.get('%scodec' % not_format_type
) == 'none')
1815 if format_type
# bv, ba, wv, wa
1816 else (lambda f
: f
.get('vcodec') != 'none' and f
.get('acodec') != 'none')
1817 if not format_modified
# b, w
1820 filter_f
= ((lambda f
: f
.get('ext') == format_spec
)
1821 if format_spec
in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1822 else (lambda f
: f
.get('format_id') == format_spec
)) # id
1824 def selector_function(ctx
):
1825 formats
= list(ctx
['formats'])
1828 matches
= list(filter(filter_f
, formats
)) if filter_f
is not None else formats
1829 if format_fallback
and ctx
['incomplete_formats'] and not matches
:
1830 # for extractors with incomplete formats (audio only (soundcloud)
1831 # or video only (imgur)) best/worst will fallback to
1832 # best/worst {video,audio}-only format
1835 matches
= matches
[::-1]
1837 matches
= list(itertools
.islice(_check_formats(matches
), format_idx
))
1839 if -n
<= format_idx
- 1 < n
:
1840 yield matches
[format_idx
- 1]
1842 elif selector
.type == MERGE
: # +
1843 selector_1
, selector_2
= map(_build_selector_function
, selector
.selector
)
1845 def selector_function(ctx
):
1846 for pair
in itertools
.product(
1847 selector_1(copy
.deepcopy(ctx
)), selector_2(copy
.deepcopy(ctx
))):
1850 filters
= [self
._build
_format
_filter
(f
) for f
in selector
.filters
]
1852 def final_selector(ctx
):
1853 ctx_copy
= copy
.deepcopy(ctx
)
1854 for _filter
in filters
:
1855 ctx_copy
['formats'] = list(filter(_filter
, ctx_copy
['formats']))
1856 return selector_function(ctx_copy
)
1857 return final_selector
1859 stream
= io
.BytesIO(format_spec
.encode('utf-8'))
1861 tokens
= list(_remove_unused_ops(compat_tokenize_tokenize(stream
.readline
)))
1862 except tokenize
.TokenError
:
1863 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec
)))
1865 class TokenIterator(object):
1866 def __init__(self
, tokens
):
1867 self
.tokens
= tokens
1874 if self
.counter
>= len(self
.tokens
):
1875 raise StopIteration()
1876 value
= self
.tokens
[self
.counter
]
1882 def restore_last_token(self
):
1885 parsed_selector
= _parse_format_selection(iter(TokenIterator(tokens
)))
1886 return _build_selector_function(parsed_selector
)
1888 def _calc_headers(self
, info_dict
):
1889 res
= std_headers
.copy()
1891 add_headers
= info_dict
.get('http_headers')
1893 res
.update(add_headers
)
1895 cookies
= self
._calc
_cookies
(info_dict
)
1897 res
['Cookie'] = cookies
1899 if 'X-Forwarded-For' not in res
:
1900 x_forwarded_for_ip
= info_dict
.get('__x_forwarded_for_ip')
1901 if x_forwarded_for_ip
:
1902 res
['X-Forwarded-For'] = x_forwarded_for_ip
1906 def _calc_cookies(self
, info_dict
):
1907 pr
= sanitized_Request(info_dict
['url'])
1908 self
.cookiejar
.add_cookie_header(pr
)
1909 return pr
.get_header('Cookie')
1912 def _sanitize_thumbnails(info_dict
):
1913 thumbnails
= info_dict
.get('thumbnails')
1914 if thumbnails
is None:
1915 thumbnail
= info_dict
.get('thumbnail')
1917 info_dict
['thumbnails'] = thumbnails
= [{'url': thumbnail}
]
1919 thumbnails
.sort(key
=lambda t
: (
1920 t
.get('preference') if t
.get('preference') is not None else -1,
1921 t
.get('width') if t
.get('width') is not None else -1,
1922 t
.get('height') if t
.get('height') is not None else -1,
1923 t
.get('id') if t
.get('id') is not None else '',
1925 for i
, t
in enumerate(thumbnails
):
1926 t
['url'] = sanitize_url(t
['url'])
1927 if t
.get('width') and t
.get('height'):
1928 t
['resolution'] = '%dx%d' % (t
['width'], t
['height'])
1929 if t
.get('id') is None:
1932 def process_video_result(self
, info_dict
, download
=True):
1933 assert info_dict
.get('_type', 'video') == 'video'
1935 if 'id' not in info_dict
:
1936 raise ExtractorError('Missing "id" field in extractor result')
1937 if 'title' not in info_dict
:
1938 raise ExtractorError('Missing "title" field in extractor result')
1940 def report_force_conversion(field
, field_not
, conversion
):
1941 self
.report_warning(
1942 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1943 % (field
, field_not
, conversion
))
1945 def sanitize_string_field(info
, string_field
):
1946 field
= info
.get(string_field
)
1947 if field
is None or isinstance(field
, compat_str
):
1949 report_force_conversion(string_field
, 'a string', 'string')
1950 info
[string_field
] = compat_str(field
)
1952 def sanitize_numeric_fields(info
):
1953 for numeric_field
in self
._NUMERIC
_FIELDS
:
1954 field
= info
.get(numeric_field
)
1955 if field
is None or isinstance(field
, compat_numeric_types
):
1957 report_force_conversion(numeric_field
, 'numeric', 'int')
1958 info
[numeric_field
] = int_or_none(field
)
1960 sanitize_string_field(info_dict
, 'id')
1961 sanitize_numeric_fields(info_dict
)
1963 if 'playlist' not in info_dict
:
1964 # It isn't part of a playlist
1965 info_dict
['playlist'] = None
1966 info_dict
['playlist_index'] = None
1968 self
._sanitize
_thumbnails
(info_dict
)
1970 if self
.params
.get('list_thumbnails'):
1971 self
.list_thumbnails(info_dict
)
1974 thumbnail
= info_dict
.get('thumbnail')
1975 thumbnails
= info_dict
.get('thumbnails')
1977 info_dict
['thumbnail'] = sanitize_url(thumbnail
)
1979 info_dict
['thumbnail'] = thumbnails
[-1]['url']
1981 if 'display_id' not in info_dict
and 'id' in info_dict
:
1982 info_dict
['display_id'] = info_dict
['id']
1984 for ts_key
, date_key
in (
1985 ('timestamp', 'upload_date'),
1986 ('release_timestamp', 'release_date'),
1988 if info_dict
.get(date_key
) is None and info_dict
.get(ts_key
) is not None:
1989 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1990 # see http://bugs.python.org/issue1646728)
1992 upload_date
= datetime
.datetime
.utcfromtimestamp(info_dict
[ts_key
])
1993 info_dict
[date_key
] = upload_date
.strftime('%Y%m%d')
1994 except (ValueError, OverflowError, OSError):
1997 # Auto generate title fields corresponding to the *_number fields when missing
1998 # in order to always have clean titles. This is very common for TV series.
1999 for field
in ('chapter', 'season', 'episode'):
2000 if info_dict
.get('%s_number' % field
) is not None and not info_dict
.get(field
):
2001 info_dict
[field
] = '%s %d' % (field
.capitalize(), info_dict
['%s_number' % field
])
2003 for cc_kind
in ('subtitles', 'automatic_captions'):
2004 cc
= info_dict
.get(cc_kind
)
2006 for _
, subtitle
in cc
.items():
2007 for subtitle_format
in subtitle
:
2008 if subtitle_format
.get('url'):
2009 subtitle_format
['url'] = sanitize_url(subtitle_format
['url'])
2010 if subtitle_format
.get('ext') is None:
2011 subtitle_format
['ext'] = determine_ext(subtitle_format
['url']).lower()
2013 automatic_captions
= info_dict
.get('automatic_captions')
2014 subtitles
= info_dict
.get('subtitles')
2016 if self
.params
.get('listsubtitles', False):
2017 if 'automatic_captions' in info_dict
:
2018 self
.list_subtitles(
2019 info_dict
['id'], automatic_captions
, 'automatic captions')
2020 self
.list_subtitles(info_dict
['id'], subtitles
, 'subtitles')
2023 info_dict
['requested_subtitles'] = self
.process_subtitles(
2024 info_dict
['id'], subtitles
, automatic_captions
)
2026 # We now pick which formats have to be downloaded
2027 if info_dict
.get('formats') is None:
2028 # There's only one format available
2029 formats
= [info_dict
]
2031 formats
= info_dict
['formats']
2034 if not self
.params
.get('ignore_no_formats_error'):
2035 raise ExtractorError('No video formats found!')
2037 self
.report_warning('No video formats found!')
2039 def is_wellformed(f
):
2042 self
.report_warning(
2043 '"url" field is missing or empty - skipping format, '
2044 'there is an error in extractor')
2046 if isinstance(url
, bytes):
2047 sanitize_string_field(f
, 'url')
2050 # Filter out malformed formats for better extraction robustness
2051 formats
= list(filter(is_wellformed
, formats
))
2055 # We check that all the formats have the format and format_id fields
2056 for i
, format
in enumerate(formats
):
2057 sanitize_string_field(format
, 'format_id')
2058 sanitize_numeric_fields(format
)
2059 format
['url'] = sanitize_url(format
['url'])
2060 if not format
.get('format_id'):
2061 format
['format_id'] = compat_str(i
)
2063 # Sanitize format_id from characters used in format selector expression
2064 format
['format_id'] = re
.sub(r
'[\s,/+\[\]()]', '_', format
['format_id'])
2065 format_id
= format
['format_id']
2066 if format_id
not in formats_dict
:
2067 formats_dict
[format_id
] = []
2068 formats_dict
[format_id
].append(format
)
2070 # Make sure all formats have unique format_id
2071 for format_id
, ambiguous_formats
in formats_dict
.items():
2072 if len(ambiguous_formats
) > 1:
2073 for i
, format
in enumerate(ambiguous_formats
):
2074 format
['format_id'] = '%s-%d' % (format_id
, i
)
2076 for i
, format
in enumerate(formats
):
2077 if format
.get('format') is None:
2078 format
['format'] = '{id} - {res}{note}'.format(
2079 id=format
['format_id'],
2080 res
=self
.format_resolution(format
),
2081 note
=' ({0})'.format(format
['format_note']) if format
.get('format_note') is not None else '',
2083 # Automatically determine file extension if missing
2084 if format
.get('ext') is None:
2085 format
['ext'] = determine_ext(format
['url']).lower()
2086 # Automatically determine protocol if missing (useful for format
2087 # selection purposes)
2088 if format
.get('protocol') is None:
2089 format
['protocol'] = determine_protocol(format
)
2090 # Add HTTP headers, so that external programs can use them from the
2092 full_format_info
= info_dict
.copy()
2093 full_format_info
.update(format
)
2094 format
['http_headers'] = self
._calc
_headers
(full_format_info
)
2095 # Remove private housekeeping stuff
2096 if '__x_forwarded_for_ip' in info_dict
:
2097 del info_dict
['__x_forwarded_for_ip']
2099 # TODO Central sorting goes here
2101 if formats
and formats
[0] is not info_dict
:
2102 # only set the 'formats' fields if the original info_dict list them
2103 # otherwise we end up with a circular reference, the first (and unique)
2104 # element in the 'formats' field in info_dict is info_dict itself,
2105 # which can't be exported to json
2106 info_dict
['formats'] = formats
2108 info_dict
, _
= self
.pre_process(info_dict
)
2110 if self
.params
.get('listformats'):
2111 if not info_dict
.get('formats'):
2112 raise ExtractorError('No video formats found', expected
=True)
2113 self
.list_formats(info_dict
)
2116 req_format
= self
.params
.get('format')
2117 if req_format
is None:
2118 req_format
= self
._default
_format
_spec
(info_dict
, download
=download
)
2119 self
.write_debug('Default format spec: %s' % req_format
)
2121 format_selector
= self
.build_format_selector(req_format
)
2123 # While in format selection we may need to have an access to the original
2124 # format set in order to calculate some metrics or do some processing.
2125 # For now we need to be able to guess whether original formats provided
2126 # by extractor are incomplete or not (i.e. whether extractor provides only
2127 # video-only or audio-only formats) for proper formats selection for
2128 # extractors with such incomplete formats (see
2129 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2130 # Since formats may be filtered during format selection and may not match
2131 # the original formats the results may be incorrect. Thus original formats
2132 # or pre-calculated metrics should be passed to format selection routines
2134 # We will pass a context object containing all necessary additional data
2135 # instead of just formats.
2136 # This fixes incorrect format selection issue (see
2137 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2138 incomplete_formats
= (
2139 # All formats are video-only or
2140 all(f
.get('vcodec') != 'none' and f
.get('acodec') == 'none' for f
in formats
)
2141 # all formats are audio-only
2142 or all(f
.get('vcodec') == 'none' and f
.get('acodec') != 'none' for f
in formats
))
2146 'incomplete_formats': incomplete_formats
,
2149 formats_to_download
= list(format_selector(ctx
))
2150 if not formats_to_download
:
2151 if not self
.params
.get('ignore_no_formats_error'):
2152 raise ExtractorError('Requested format is not available', expected
=True)
2154 self
.report_warning('Requested format is not available')
2157 '[info] %s: Downloading %d format(s): %s' % (
2158 info_dict
['id'], len(formats_to_download
),
2159 ", ".join([f
['format_id'] for f
in formats_to_download
])))
2160 for fmt
in formats_to_download
:
2161 new_info
= dict(info_dict
)
2162 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2163 new_info
['__original_infodict'] = info_dict
2164 new_info
.update(fmt
)
2165 self
.process_info(new_info
)
2166 # We update the info dict with the best quality format (backwards compatibility)
2167 if formats_to_download
:
2168 info_dict
.update(formats_to_download
[-1])
2171 def process_subtitles(self
, video_id
, normal_subtitles
, automatic_captions
):
2172 """Select the requested subtitles and their format"""
2174 if normal_subtitles
and self
.params
.get('writesubtitles'):
2175 available_subs
.update(normal_subtitles
)
2176 if automatic_captions
and self
.params
.get('writeautomaticsub'):
2177 for lang
, cap_info
in automatic_captions
.items():
2178 if lang
not in available_subs
:
2179 available_subs
[lang
] = cap_info
2181 if (not self
.params
.get('writesubtitles') and not
2182 self
.params
.get('writeautomaticsub') or not
2186 all_sub_langs
= available_subs
.keys()
2187 if self
.params
.get('allsubtitles', False):
2188 requested_langs
= all_sub_langs
2189 elif self
.params
.get('subtitleslangs', False):
2190 requested_langs
= set()
2191 for lang
in self
.params
.get('subtitleslangs'):
2193 requested_langs
.update(all_sub_langs
)
2195 discard
= lang
[0] == '-'
2198 current_langs
= filter(re
.compile(lang
+ '$').match
, all_sub_langs
)
2200 for lang
in current_langs
:
2201 requested_langs
.discard(lang
)
2203 requested_langs
.update(current_langs
)
2204 elif 'en' in available_subs
:
2205 requested_langs
= ['en']
2207 requested_langs
= [list(all_sub_langs
)[0]]
2208 self
.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs
))
2210 formats_query
= self
.params
.get('subtitlesformat', 'best')
2211 formats_preference
= formats_query
.split('/') if formats_query
else []
2213 for lang
in requested_langs
:
2214 formats
= available_subs
.get(lang
)
2216 self
.report_warning('%s subtitles not available for %s' % (lang
, video_id
))
2218 for ext
in formats_preference
:
2222 matches
= list(filter(lambda f
: f
['ext'] == ext
, formats
))
2228 self
.report_warning(
2229 'No subtitle format found matching "%s" for language %s, '
2230 'using %s' % (formats_query
, lang
, f
['ext']))
2234 def __forced_printings(self
, info_dict
, filename
, incomplete
):
2235 def print_mandatory(field
, actual_field
=None):
2236 if actual_field
is None:
2237 actual_field
= field
2238 if (self
.params
.get('force%s' % field
, False)
2239 and (not incomplete
or info_dict
.get(actual_field
) is not None)):
2240 self
.to_stdout(info_dict
[actual_field
])
2242 def print_optional(field
):
2243 if (self
.params
.get('force%s' % field
, False)
2244 and info_dict
.get(field
) is not None):
2245 self
.to_stdout(info_dict
[field
])
2247 info_dict
= info_dict
.copy()
2248 if filename
is not None:
2249 info_dict
['filename'] = filename
2250 if info_dict
.get('requested_formats') is not None:
2251 # For RTMP URLs, also include the playpath
2252 info_dict
['urls'] = '\n'.join(f
['url'] + f
.get('play_path', '') for f
in info_dict
['requested_formats'])
2253 elif 'url' in info_dict
:
2254 info_dict
['urls'] = info_dict
['url'] + info_dict
.get('play_path', '')
2256 for tmpl
in self
.params
.get('forceprint', []):
2257 if re
.match(r
'\w+$', tmpl
):
2258 tmpl
= '%({})s'.format(tmpl
)
2259 tmpl
, info_copy
= self
.prepare_outtmpl(tmpl
, info_dict
)
2260 self
.to_stdout(tmpl
% info_copy
)
2262 print_mandatory('title')
2263 print_mandatory('id')
2264 print_mandatory('url', 'urls')
2265 print_optional('thumbnail')
2266 print_optional('description')
2267 print_optional('filename')
2268 if self
.params
.get('forceduration', False) and info_dict
.get('duration') is not None:
2269 self
.to_stdout(formatSeconds(info_dict
['duration']))
2270 print_mandatory('format')
2272 if self
.params
.get('forcejson', False):
2273 self
.post_extract(info_dict
)
2274 self
.to_stdout(json
.dumps(info_dict
, default
=repr))
2276 def dl(self
, name
, info
, subtitle
=False, test
=False):
2279 verbose
= self
.params
.get('verbose')
2282 'quiet': not verbose
,
2284 'noprogress': not verbose
,
2286 'skip_unavailable_fragments': False,
2287 'keep_fragments': False,
2289 '_no_ytdl_file': True,
2292 params
= self
.params
2293 fd
= get_suitable_downloader(info
, params
)(self
, params
)
2295 for ph
in self
._progress
_hooks
:
2296 fd
.add_progress_hook(ph
)
2297 urls
= '", "'.join([f
['url'] for f
in info
.get('requested_formats', [])] or [info
['url']])
2298 self
.write_debug('Invoking downloader on "%s"' % urls
)
2299 new_info
= dict(info
)
2300 if new_info
.get('http_headers') is None:
2301 new_info
['http_headers'] = self
._calc
_headers
(new_info
)
2302 return fd
.download(name
, new_info
, subtitle
)
2304 def process_info(self
, info_dict
):
2305 """Process a single resolved IE result."""
2307 assert info_dict
.get('_type', 'video') == 'video'
2309 info_dict
.setdefault('__postprocessors', [])
2311 max_downloads
= self
.params
.get('max_downloads')
2312 if max_downloads
is not None:
2313 if self
._num
_downloads
>= int(max_downloads
):
2314 raise MaxDownloadsReached()
2316 # TODO: backward compatibility, to be removed
2317 info_dict
['fulltitle'] = info_dict
['title']
2319 if 'format' not in info_dict
:
2320 info_dict
['format'] = info_dict
['ext']
2322 if self
._match
_entry
(info_dict
) is not None:
2325 self
.post_extract(info_dict
)
2326 self
._num
_downloads
+= 1
2328 # info_dict['_filename'] needs to be set for backward compatibility
2329 info_dict
['_filename'] = full_filename
= self
.prepare_filename(info_dict
, warn
=True)
2330 temp_filename
= self
.prepare_filename(info_dict
, 'temp')
2334 self
.__forced
_printings
(info_dict
, full_filename
, incomplete
=False)
2336 if self
.params
.get('simulate', False):
2337 if self
.params
.get('force_write_download_archive', False):
2338 self
.record_download_archive(info_dict
)
2340 # Do nothing else if in simulate mode
2343 if full_filename
is None:
2346 if not self
._ensure
_dir
_exists
(encodeFilename(full_filename
)):
2348 if not self
._ensure
_dir
_exists
(encodeFilename(temp_filename
)):
2351 if self
.params
.get('writedescription', False):
2352 descfn
= self
.prepare_filename(info_dict
, 'description')
2353 if not self
._ensure
_dir
_exists
(encodeFilename(descfn
)):
2355 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(descfn
)):
2356 self
.to_screen('[info] Video description is already present')
2357 elif info_dict
.get('description') is None:
2358 self
.report_warning('There\'s no description to write.')
2361 self
.to_screen('[info] Writing video description to: ' + descfn
)
2362 with io
.open(encodeFilename(descfn
), 'w', encoding
='utf-8') as descfile
:
2363 descfile
.write(info_dict
['description'])
2364 except (OSError, IOError):
2365 self
.report_error('Cannot write description file ' + descfn
)
2368 if self
.params
.get('writeannotations', False):
2369 annofn
= self
.prepare_filename(info_dict
, 'annotation')
2370 if not self
._ensure
_dir
_exists
(encodeFilename(annofn
)):
2372 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(annofn
)):
2373 self
.to_screen('[info] Video annotations are already present')
2374 elif not info_dict
.get('annotations'):
2375 self
.report_warning('There are no annotations to write.')
2378 self
.to_screen('[info] Writing video annotations to: ' + annofn
)
2379 with io
.open(encodeFilename(annofn
), 'w', encoding
='utf-8') as annofile
:
2380 annofile
.write(info_dict
['annotations'])
2381 except (KeyError, TypeError):
2382 self
.report_warning('There are no annotations to write.')
2383 except (OSError, IOError):
2384 self
.report_error('Cannot write annotations file: ' + annofn
)
2387 subtitles_are_requested
= any([self
.params
.get('writesubtitles', False),
2388 self
.params
.get('writeautomaticsub')])
2390 if subtitles_are_requested
and info_dict
.get('requested_subtitles'):
2391 # subtitles download errors are already managed as troubles in relevant IE
2392 # that way it will silently go on when used with unsupporting IE
2393 subtitles
= info_dict
['requested_subtitles']
2394 # ie = self.get_info_extractor(info_dict['extractor_key'])
2395 for sub_lang
, sub_info
in subtitles
.items():
2396 sub_format
= sub_info
['ext']
2397 sub_filename
= subtitles_filename(temp_filename
, sub_lang
, sub_format
, info_dict
.get('ext'))
2398 sub_filename_final
= subtitles_filename(
2399 self
.prepare_filename(info_dict
, 'subtitle'), sub_lang
, sub_format
, info_dict
.get('ext'))
2400 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(sub_filename
)):
2401 self
.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang
, sub_format
))
2402 sub_info
['filepath'] = sub_filename
2403 files_to_move
[sub_filename
] = sub_filename_final
2405 self
.to_screen('[info] Writing video subtitles to: ' + sub_filename
)
2406 if sub_info
.get('data') is not None:
2408 # Use newline='' to prevent conversion of newline characters
2409 # See https://github.com/ytdl-org/youtube-dl/issues/10268
2410 with io
.open(encodeFilename(sub_filename
), 'w', encoding
='utf-8', newline
='') as subfile
:
2411 subfile
.write(sub_info
['data'])
2412 sub_info
['filepath'] = sub_filename
2413 files_to_move
[sub_filename
] = sub_filename_final
2414 except (OSError, IOError):
2415 self
.report_error('Cannot write subtitles file ' + sub_filename
)
2419 self
.dl(sub_filename
, sub_info
.copy(), subtitle
=True)
2420 sub_info
['filepath'] = sub_filename
2421 files_to_move
[sub_filename
] = sub_filename_final
2422 except (ExtractorError
, IOError, OSError, ValueError) + network_exceptions
as err
:
2423 self
.report_warning('Unable to download subtitle for "%s": %s' %
2424 (sub_lang
, error_to_compat_str(err
)))
2427 if self
.params
.get('writeinfojson', False):
2428 infofn
= self
.prepare_filename(info_dict
, 'infojson')
2429 if not self
._ensure
_dir
_exists
(encodeFilename(infofn
)):
2431 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(infofn
)):
2432 self
.to_screen('[info] Video metadata is already present')
2434 self
.to_screen('[info] Writing video metadata as JSON to: ' + infofn
)
2436 write_json_file(self
.filter_requested_info(info_dict
, self
.params
.get('clean_infojson', True)), infofn
)
2437 except (OSError, IOError):
2438 self
.report_error('Cannot write video metadata to JSON file ' + infofn
)
2440 info_dict
['__infojson_filename'] = infofn
2442 for thumb_ext
in self
._write
_thumbnails
(info_dict
, temp_filename
):
2443 thumb_filename_temp
= replace_extension(temp_filename
, thumb_ext
, info_dict
.get('ext'))
2444 thumb_filename
= replace_extension(
2445 self
.prepare_filename(info_dict
, 'thumbnail'), thumb_ext
, info_dict
.get('ext'))
2446 files_to_move
[thumb_filename_temp
] = thumb_filename
2448 # Write internet shortcut files
2449 url_link
= webloc_link
= desktop_link
= False
2450 if self
.params
.get('writelink', False):
2451 if sys
.platform
== "darwin": # macOS.
2453 elif sys
.platform
.startswith("linux"):
2455 else: # if sys.platform in ['win32', 'cygwin']:
2457 if self
.params
.get('writeurllink', False):
2459 if self
.params
.get('writewebloclink', False):
2461 if self
.params
.get('writedesktoplink', False):
2464 if url_link
or webloc_link
or desktop_link
:
2465 if 'webpage_url' not in info_dict
:
2466 self
.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2468 ascii_url
= iri_to_uri(info_dict
['webpage_url'])
2470 def _write_link_file(extension
, template
, newline
, embed_filename
):
2471 linkfn
= replace_extension(full_filename
, extension
, info_dict
.get('ext'))
2472 if self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(linkfn
)):
2473 self
.to_screen('[info] Internet shortcut is already present')
2476 self
.to_screen('[info] Writing internet shortcut to: ' + linkfn
)
2477 with io
.open(encodeFilename(to_high_limit_path(linkfn
)), 'w', encoding
='utf-8', newline
=newline
) as linkfile
:
2478 template_vars
= {'url': ascii_url}
2480 template_vars
['filename'] = linkfn
[:-(len(extension
) + 1)]
2481 linkfile
.write(template
% template_vars
)
2482 except (OSError, IOError):
2483 self
.report_error('Cannot write internet shortcut ' + linkfn
)
2488 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE
, '\r\n', embed_filename
=False):
2491 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE
, '\n', embed_filename
=False):
2494 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE
, '\n', embed_filename
=True):
2498 info_dict
, files_to_move
= self
.pre_process(info_dict
, 'before_dl', files_to_move
)
2499 except PostProcessingError
as err
:
2500 self
.report_error('Preprocessing: %s' % str(err
))
2503 must_record_download_archive
= False
2504 if self
.params
.get('skip_download', False):
2505 info_dict
['filepath'] = temp_filename
2506 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2507 info_dict
['__files_to_move'] = files_to_move
2508 info_dict
= self
.run_pp(MoveFilesAfterDownloadPP(self
, False), info_dict
)
2513 def existing_file(*filepaths
):
2514 ext
= info_dict
.get('ext')
2515 final_ext
= self
.params
.get('final_ext', ext
)
2517 for file in orderedSet(filepaths
):
2518 if final_ext
!= ext
:
2519 converted
= replace_extension(file, final_ext
, ext
)
2520 if os
.path
.exists(encodeFilename(converted
)):
2521 existing_files
.append(converted
)
2522 if os
.path
.exists(encodeFilename(file)):
2523 existing_files
.append(file)
2525 if not existing_files
or self
.params
.get('overwrites', False):
2526 for file in orderedSet(existing_files
):
2527 self
.report_file_delete(file)
2528 os
.remove(encodeFilename(file))
2531 self
.report_file_already_downloaded(existing_files
[0])
2532 info_dict
['ext'] = os
.path
.splitext(existing_files
[0])[1][1:]
2533 return existing_files
[0]
2536 if info_dict
.get('requested_formats') is not None:
2538 def compatible_formats(formats
):
2539 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2540 video_formats
= [format
for format
in formats
if format
.get('vcodec') != 'none']
2541 audio_formats
= [format
for format
in formats
if format
.get('acodec') != 'none']
2542 if len(video_formats
) > 2 or len(audio_formats
) > 2:
2546 exts
= set(format
.get('ext') for format
in formats
)
2548 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2551 for ext_sets
in COMPATIBLE_EXTS
:
2552 if ext_sets
.issuperset(exts
):
2554 # TODO: Check acodec/vcodec
2557 requested_formats
= info_dict
['requested_formats']
2558 old_ext
= info_dict
['ext']
2559 if self
.params
.get('merge_output_format') is None:
2560 if not compatible_formats(requested_formats
):
2561 info_dict
['ext'] = 'mkv'
2562 self
.report_warning(
2563 'Requested formats are incompatible for merge and will be merged into mkv.')
2564 if (info_dict
['ext'] == 'webm'
2565 and self
.params
.get('writethumbnail', False)
2566 and info_dict
.get('thumbnails')):
2567 info_dict
['ext'] = 'mkv'
2568 self
.report_warning(
2569 'webm doesn\'t support embedding a thumbnail, mkv will be used.')
2571 def correct_ext(filename
):
2572 filename_real_ext
= os
.path
.splitext(filename
)[1][1:]
2574 os
.path
.splitext(filename
)[0]
2575 if filename_real_ext
== old_ext
2577 return '%s.%s' % (filename_wo_ext
, info_dict
['ext'])
2579 # Ensure filename always has a correct extension for successful merge
2580 full_filename
= correct_ext(full_filename
)
2581 temp_filename
= correct_ext(temp_filename
)
2582 dl_filename
= existing_file(full_filename
, temp_filename
)
2583 info_dict
['__real_download'] = False
2585 _protocols
= set(determine_protocol(f
) for f
in requested_formats
)
2586 if len(_protocols
) == 1:
2587 info_dict
['protocol'] = _protocols
.pop()
2588 directly_mergable
= (
2589 'no-direct-merge' not in self
.params
.get('compat_opts', [])
2590 and info_dict
.get('protocol') is not None # All requested formats have same protocol
2591 and not self
.params
.get('allow_unplayable_formats')
2592 and get_suitable_downloader(info_dict
, self
.params
).__name
__ == 'FFmpegFD')
2593 if directly_mergable
:
2594 info_dict
['url'] = requested_formats
[0]['url']
2595 # Treat it as a single download
2596 dl_filename
= existing_file(full_filename
, temp_filename
)
2597 if dl_filename
is None:
2598 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2599 info_dict
['__real_download'] = real_download
2602 merger
= FFmpegMergerPP(self
)
2603 if self
.params
.get('allow_unplayable_formats'):
2604 self
.report_warning(
2605 'You have requested merging of multiple formats '
2606 'while also allowing unplayable formats to be downloaded. '
2607 'The formats won\'t be merged to prevent data corruption.')
2608 elif not merger
.available
:
2609 self
.report_warning(
2610 'You have requested merging of multiple formats but ffmpeg is not installed. '
2611 'The formats won\'t be merged.')
2613 if dl_filename
is None:
2614 for f
in requested_formats
:
2615 new_info
= dict(info_dict
)
2616 del new_info
['requested_formats']
2618 fname
= prepend_extension(
2619 self
.prepare_filename(new_info
, 'temp'),
2620 'f%s' % f
['format_id'], new_info
['ext'])
2621 if not self
._ensure
_dir
_exists
(fname
):
2623 downloaded
.append(fname
)
2624 partial_success
, real_download
= self
.dl(fname
, new_info
)
2625 info_dict
['__real_download'] = info_dict
['__real_download'] or real_download
2626 success
= success
and partial_success
2627 if merger
.available
and not self
.params
.get('allow_unplayable_formats'):
2628 info_dict
['__postprocessors'].append(merger
)
2629 info_dict
['__files_to_merge'] = downloaded
2630 # Even if there were no downloads, it is being merged only now
2631 info_dict
['__real_download'] = True
2633 for file in downloaded
:
2634 files_to_move
[file] = None
2636 # Just a single file
2637 dl_filename
= existing_file(full_filename
, temp_filename
)
2638 if dl_filename
is None:
2639 success
, real_download
= self
.dl(temp_filename
, info_dict
)
2640 info_dict
['__real_download'] = real_download
2642 dl_filename
= dl_filename
or temp_filename
2643 info_dict
['__finaldir'] = os
.path
.dirname(os
.path
.abspath(encodeFilename(full_filename
)))
2645 except network_exceptions
as err
:
2646 self
.report_error('unable to download video data: %s' % error_to_compat_str(err
))
2648 except (OSError, IOError) as err
:
2649 raise UnavailableVideoError(err
)
2650 except (ContentTooShortError
, ) as err
:
2651 self
.report_error('content too short (expected %s bytes and served %s)' % (err
.expected
, err
.downloaded
))
2654 if success
and full_filename
!= '-':
2656 fixup_policy
= self
.params
.get('fixup')
2657 if fixup_policy
is None:
2658 fixup_policy
= 'detect_or_warn'
2660 INSTALL_FFMPEG_MESSAGE
= 'Install ffmpeg to fix this automatically.'
2662 stretched_ratio
= info_dict
.get('stretched_ratio')
2663 if stretched_ratio
is not None and stretched_ratio
!= 1:
2664 if fixup_policy
== 'warn':
2665 self
.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2666 info_dict
['id'], stretched_ratio
))
2667 elif fixup_policy
== 'detect_or_warn':
2668 stretched_pp
= FFmpegFixupStretchedPP(self
)
2669 if stretched_pp
.available
:
2670 info_dict
['__postprocessors'].append(stretched_pp
)
2672 self
.report_warning(
2673 '%s: Non-uniform pixel ratio (%s). %s'
2674 % (info_dict
['id'], stretched_ratio
, INSTALL_FFMPEG_MESSAGE
))
2676 assert fixup_policy
in ('ignore', 'never')
2678 if (info_dict
.get('requested_formats') is None
2679 and info_dict
.get('container') == 'm4a_dash'
2680 and info_dict
.get('ext') == 'm4a'):
2681 if fixup_policy
== 'warn':
2682 self
.report_warning(
2683 '%s: writing DASH m4a. '
2684 'Only some players support this container.'
2686 elif fixup_policy
== 'detect_or_warn':
2687 fixup_pp
= FFmpegFixupM4aPP(self
)
2688 if fixup_pp
.available
:
2689 info_dict
['__postprocessors'].append(fixup_pp
)
2691 self
.report_warning(
2692 '%s: writing DASH m4a. '
2693 'Only some players support this container. %s'
2694 % (info_dict
['id'], INSTALL_FFMPEG_MESSAGE
))
2696 assert fixup_policy
in ('ignore', 'never')
2698 if ('protocol' in info_dict
2699 and get_suitable_downloader(info_dict
, self
.params
).__name
__ == 'HlsFD'):
2700 if fixup_policy
== 'warn':
2701 self
.report_warning('%s: malformed AAC bitstream detected.' % (
2703 elif fixup_policy
== 'detect_or_warn':
2704 fixup_pp
= FFmpegFixupM3u8PP(self
)
2705 if fixup_pp
.available
:
2706 info_dict
['__postprocessors'].append(fixup_pp
)
2708 self
.report_warning(
2709 '%s: malformed AAC bitstream detected. %s'
2710 % (info_dict
['id'], INSTALL_FFMPEG_MESSAGE
))
2712 assert fixup_policy
in ('ignore', 'never')
2715 info_dict
= self
.post_process(dl_filename
, info_dict
, files_to_move
)
2716 except PostProcessingError
as err
:
2717 self
.report_error('Postprocessing: %s' % str(err
))
2720 for ph
in self
._post
_hooks
:
2721 ph(info_dict
['filepath'])
2722 except Exception as err
:
2723 self
.report_error('post hooks: %s' % str(err
))
2725 must_record_download_archive
= True
2727 if must_record_download_archive
or self
.params
.get('force_write_download_archive', False):
2728 self
.record_download_archive(info_dict
)
2729 max_downloads
= self
.params
.get('max_downloads')
2730 if max_downloads
is not None and self
._num
_downloads
>= int(max_downloads
):
2731 raise MaxDownloadsReached()
2733 def download(self
, url_list
):
2734 """Download a given list of URLs."""
2735 outtmpl
= self
.outtmpl_dict
['default']
2736 if (len(url_list
) > 1
2738 and '%' not in outtmpl
2739 and self
.params
.get('max_downloads') != 1):
2740 raise SameFileError(outtmpl
)
2742 for url
in url_list
:
2744 # It also downloads the videos
2745 res
= self
.extract_info(
2746 url
, force_generic_extractor
=self
.params
.get('force_generic_extractor', False))
2747 except UnavailableVideoError
:
2748 self
.report_error('unable to download video')
2749 except MaxDownloadsReached
:
2750 self
.to_screen('[info] Maximum number of downloaded files reached')
2752 except ExistingVideoReached
:
2753 self
.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
2755 except RejectedVideoReached
:
2756 self
.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
2759 if self
.params
.get('dump_single_json', False):
2760 self
.post_extract(res
)
2761 self
.to_stdout(json
.dumps(res
, default
=repr))
2763 return self
._download
_retcode
2765 def download_with_info_file(self
, info_filename
):
2766 with contextlib
.closing(fileinput
.FileInput(
2767 [info_filename
], mode
='r',
2768 openhook
=fileinput
.hook_encoded('utf-8'))) as f
:
2769 # FileInput doesn't have a read method, we can't call json.load
2770 info
= self
.filter_requested_info(json
.loads('\n'.join(f
)), self
.params
.get('clean_infojson', True))
2772 self
.process_ie_result(info
, download
=True)
2773 except (DownloadError
, EntryNotInPlaylist
):
2774 webpage_url
= info
.get('webpage_url')
2775 if webpage_url
is not None:
2776 self
.report_warning('The info failed to download, trying with "%s"' % webpage_url
)
2777 return self
.download([webpage_url
])
2780 return self
._download
_retcode
2783 def filter_requested_info(info_dict
, actually_filter
=True):
2784 remove_keys
= ['__original_infodict'] # Always remove this since this may contain a copy of the entire dict
2785 keep_keys
= ['_type'], # Always keep this to facilitate load-info-json
2787 remove_keys
+= ('requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries', 'original_url')
2788 empty_values
= (None, {}, [], set(), tuple())
2789 reject
= lambda k
, v
: k
not in keep_keys
and (
2790 k
.startswith('_') or k
in remove_keys
or v
in empty_values
)
2792 info_dict
['epoch'] = int(time
.time())
2793 reject
= lambda k
, v
: k
in remove_keys
2794 filter_fn
= lambda obj
: (
2795 list(map(filter_fn
, obj
)) if isinstance(obj
, (list, tuple, set))
2796 else obj
if not isinstance(obj
, dict)
2797 else dict((k
, filter_fn(v
)) for k
, v
in obj
.items() if not reject(k
, v
)))
2798 return filter_fn(info_dict
)
2800 def run_pp(self
, pp
, infodict
):
2801 files_to_delete
= []
2802 if '__files_to_move' not in infodict
:
2803 infodict
['__files_to_move'] = {}
2804 files_to_delete
, infodict
= pp
.run(infodict
)
2805 if not files_to_delete
:
2808 if self
.params
.get('keepvideo', False):
2809 for f
in files_to_delete
:
2810 infodict
['__files_to_move'].setdefault(f
, '')
2812 for old_filename
in set(files_to_delete
):
2813 self
.to_screen('Deleting original file %s (pass -k to keep)' % old_filename
)
2815 os
.remove(encodeFilename(old_filename
))
2816 except (IOError, OSError):
2817 self
.report_warning('Unable to remove downloaded original file')
2818 if old_filename
in infodict
['__files_to_move']:
2819 del infodict
['__files_to_move'][old_filename
]
2823 def post_extract(info_dict
):
2824 def actual_post_extract(info_dict
):
2825 if info_dict
.get('_type') in ('playlist', 'multi_video'):
2826 for video_dict
in info_dict
.get('entries', {}):
2827 actual_post_extract(video_dict
or {})
2830 post_extractor
= info_dict
.get('__post_extractor') or (lambda: {})
2831 extra
= post_extractor().items()
2832 info_dict
.update(extra
)
2833 info_dict
.pop('__post_extractor', None)
2835 original_infodict
= info_dict
.get('__original_infodict') or {}
2836 original_infodict
.update(extra
)
2837 original_infodict
.pop('__post_extractor', None)
2839 actual_post_extract(info_dict
or {})
2841 def pre_process(self
, ie_info
, key
='pre_process', files_to_move
=None):
2842 info
= dict(ie_info
)
2843 info
['__files_to_move'] = files_to_move
or {}
2844 for pp
in self
._pps
[key
]:
2845 info
= self
.run_pp(pp
, info
)
2846 return info
, info
.pop('__files_to_move', None)
2848 def post_process(self
, filename
, ie_info
, files_to_move
=None):
2849 """Run all the postprocessors on the given file."""
2850 info
= dict(ie_info
)
2851 info
['filepath'] = filename
2852 info
['__files_to_move'] = files_to_move
or {}
2854 for pp
in ie_info
.get('__postprocessors', []) + self
._pps
['post_process']:
2855 info
= self
.run_pp(pp
, info
)
2856 info
= self
.run_pp(MoveFilesAfterDownloadPP(self
), info
)
2857 del info
['__files_to_move']
2858 for pp
in self
._pps
['after_move']:
2859 info
= self
.run_pp(pp
, info
)
2862 def _make_archive_id(self
, info_dict
):
2863 video_id
= info_dict
.get('id')
2866 # Future-proof against any change in case
2867 # and backwards compatibility with prior versions
2868 extractor
= info_dict
.get('extractor_key') or info_dict
.get('ie_key') # key in a playlist
2869 if extractor
is None:
2870 url
= str_or_none(info_dict
.get('url'))
2873 # Try to find matching extractor for the URL and take its ie_key
2874 for ie
in self
._ies
:
2875 if ie
.suitable(url
):
2876 extractor
= ie
.ie_key()
2880 return '%s %s' % (extractor
.lower(), video_id
)
2882 def in_download_archive(self
, info_dict
):
2883 fn
= self
.params
.get('download_archive')
2887 vid_id
= self
._make
_archive
_id
(info_dict
)
2889 return False # Incomplete video information
2891 return vid_id
in self
.archive
2893 def record_download_archive(self
, info_dict
):
2894 fn
= self
.params
.get('download_archive')
2897 vid_id
= self
._make
_archive
_id
(info_dict
)
2899 with locked_file(fn
, 'a', encoding
='utf-8') as archive_file
:
2900 archive_file
.write(vid_id
+ '\n')
2901 self
.archive
.add(vid_id
)
2904 def format_resolution(format
, default
='unknown'):
2905 if format
.get('vcodec') == 'none':
2907 if format
.get('resolution') is not None:
2908 return format
['resolution']
2909 if format
.get('width') and format
.get('height'):
2910 res
= '%dx%d' % (format
['width'], format
['height'])
2911 elif format
.get('height'):
2912 res
= '%sp' % format
['height']
2913 elif format
.get('width'):
2914 res
= '%dx?' % format
['width']
2919 def _format_note(self
, fdict
):
2921 if fdict
.get('ext') in ['f4f', 'f4m']:
2922 res
+= '(unsupported) '
2923 if fdict
.get('language'):
2926 res
+= '[%s] ' % fdict
['language']
2927 if fdict
.get('format_note') is not None:
2928 res
+= fdict
['format_note'] + ' '
2929 if fdict
.get('tbr') is not None:
2930 res
+= '%4dk ' % fdict
['tbr']
2931 if fdict
.get('container') is not None:
2934 res
+= '%s container' % fdict
['container']
2935 if (fdict
.get('vcodec') is not None
2936 and fdict
.get('vcodec') != 'none'):
2939 res
+= fdict
['vcodec']
2940 if fdict
.get('vbr') is not None:
2942 elif fdict
.get('vbr') is not None and fdict
.get('abr') is not None:
2944 if fdict
.get('vbr') is not None:
2945 res
+= '%4dk' % fdict
['vbr']
2946 if fdict
.get('fps') is not None:
2949 res
+= '%sfps' % fdict
['fps']
2950 if fdict
.get('acodec') is not None:
2953 if fdict
['acodec'] == 'none':
2956 res
+= '%-5s' % fdict
['acodec']
2957 elif fdict
.get('abr') is not None:
2961 if fdict
.get('abr') is not None:
2962 res
+= '@%3dk' % fdict
['abr']
2963 if fdict
.get('asr') is not None:
2964 res
+= ' (%5dHz)' % fdict
['asr']
2965 if fdict
.get('filesize') is not None:
2968 res
+= format_bytes(fdict
['filesize'])
2969 elif fdict
.get('filesize_approx') is not None:
2972 res
+= '~' + format_bytes(fdict
['filesize_approx'])
2975 def _format_note_table(self
, f
):
2976 def join_fields(*vargs
):
2977 return ', '.join((val
for val
in vargs
if val
!= ''))
2980 'UNSUPPORTED' if f
.get('ext') in ('f4f', 'f4m') else '',
2981 format_field(f
, 'language', '[%s]'),
2982 format_field(f
, 'format_note'),
2983 format_field(f
, 'container', ignore
=(None, f
.get('ext'))),
2984 format_field(f
, 'asr', '%5dHz'))
2986 def list_formats(self
, info_dict
):
2987 formats
= info_dict
.get('formats', [info_dict
])
2989 'list-formats' not in self
.params
.get('compat_opts', [])
2990 and self
.params
.get('list_formats_as_table', True) is not False)
2994 format_field(f
, 'format_id'),
2995 format_field(f
, 'ext'),
2996 self
.format_resolution(f
),
2997 format_field(f
, 'fps', '%d'),
2999 format_field(f
, 'filesize', ' %s', func
=format_bytes
) + format_field(f
, 'filesize_approx', '~%s', func
=format_bytes
),
3000 format_field(f
, 'tbr', '%4dk'),
3001 shorten_protocol_name(f
.get('protocol', '').replace("native", "n")),
3003 format_field(f
, 'vcodec', default
='unknown').replace('none', ''),
3004 format_field(f
, 'vbr', '%4dk'),
3005 format_field(f
, 'acodec', default
='unknown').replace('none', ''),
3006 format_field(f
, 'abr', '%3dk'),
3007 format_field(f
, 'asr', '%5dHz'),
3008 self
._format
_note
_table
(f
)]
3010 if f
.get('preference') is None or f
['preference'] >= -1000]
3011 header_line
= ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3012 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
3016 format_field(f
, 'format_id'),
3017 format_field(f
, 'ext'),
3018 self
.format_resolution(f
),
3019 self
._format
_note
(f
)]
3021 if f
.get('preference') is None or f
['preference'] >= -1000]
3022 header_line
= ['format code', 'extension', 'resolution', 'note']
3025 '[info] Available formats for %s:\n%s' % (info_dict
['id'], render_table(
3029 extraGap
=(0 if new_format
else 1),
3030 hideEmpty
=new_format
)))
3032 def list_thumbnails(self
, info_dict
):
3033 thumbnails
= info_dict
.get('thumbnails')
3035 self
.to_screen('[info] No thumbnails present for %s' % info_dict
['id'])
3039 '[info] Thumbnails for %s:' % info_dict
['id'])
3040 self
.to_screen(render_table(
3041 ['ID', 'width', 'height', 'URL'],
3042 [[t
['id'], t
.get('width', 'unknown'), t
.get('height', 'unknown'), t
['url']] for t
in thumbnails
]))
3044 def list_subtitles(self
, video_id
, subtitles
, name
='subtitles'):
3046 self
.to_screen('%s has no %s' % (video_id
, name
))
3049 'Available %s for %s:' % (name
, video_id
))
3051 def _row(lang
, formats
):
3052 exts
, names
= zip(*((f
['ext'], f
.get('name', 'unknown')) for f
in reversed(formats
)))
3053 if len(set(names
)) == 1:
3054 names
= [] if names
[0] == 'unknown' else names
[:1]
3055 return [lang
, ', '.join(names
), ', '.join(exts
)]
3057 self
.to_screen(render_table(
3058 ['Language', 'Name', 'Formats'],
3059 [_row(lang
, formats
) for lang
, formats
in subtitles
.items()],
3062 def urlopen(self
, req
):
3063 """ Start an HTTP download """
3064 if isinstance(req
, compat_basestring
):
3065 req
= sanitized_Request(req
)
3066 return self
._opener
.open(req
, timeout
=self
._socket
_timeout
)
3068 def print_debug_header(self
):
3069 if not self
.params
.get('verbose'):
3072 if type('') is not compat_str
:
3073 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
3074 self
.report_warning(
3075 'Your Python is broken! Update to a newer and supported version')
3077 stdout_encoding
= getattr(
3078 sys
.stdout
, 'encoding', 'missing (%s)' % type(sys
.stdout
).__name
__)
3080 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3081 locale
.getpreferredencoding(),
3082 sys
.getfilesystemencoding(),
3084 self
.get_encoding()))
3085 write_string(encoding_str
, encoding
=None)
3088 '(exe)' if hasattr(sys
, 'frozen')
3089 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter
)
3090 else '(source)' if os
.path
.basename(sys
.argv
[0]) == '__main__.py'
3092 self
._write
_string
('[debug] yt-dlp version %s %s\n' % (__version__
, source
))
3094 self
._write
_string
('[debug] Lazy loading extractors enabled\n')
3097 '[debug] Plugin Extractors: %s\n' % [ie
.ie_key() for ie
in _PLUGIN_CLASSES
])
3098 if self
.params
.get('compat_opts'):
3100 '[debug] Compatibility options: %s\n' % ', '.join(self
.params
.get('compat_opts')))
3102 sp
= subprocess
.Popen(
3103 ['git', 'rev-parse', '--short', 'HEAD'],
3104 stdout
=subprocess
.PIPE
, stderr
=subprocess
.PIPE
,
3105 cwd
=os
.path
.dirname(os
.path
.abspath(__file__
)))
3106 out
, err
= process_communicate_or_kill(sp
)
3107 out
= out
.decode().strip()
3108 if re
.match('[0-9a-f]+', out
):
3109 self
._write
_string
('[debug] Git HEAD: %s\n' % out
)
3116 def python_implementation():
3117 impl_name
= platform
.python_implementation()
3118 if impl_name
== 'PyPy' and hasattr(sys
, 'pypy_version_info'):
3119 return impl_name
+ ' version %d.%d.%d' % sys
.pypy_version_info
[:3]
3122 self
._write
_string
('[debug] Python version %s (%s %s) - %s\n' % (
3123 platform
.python_version(),
3124 python_implementation(),
3125 platform
.architecture()[0],
3128 exe_versions
= FFmpegPostProcessor
.get_versions(self
)
3129 exe_versions
['rtmpdump'] = rtmpdump_version()
3130 exe_versions
['phantomjs'] = PhantomJSwrapper
._version
()
3131 exe_str
= ', '.join(
3133 for exe
, v
in sorted(exe_versions
.items())
3138 self
._write
_string
('[debug] exe versions: %s\n' % exe_str
)
3141 for handler
in self
._opener
.handlers
:
3142 if hasattr(handler
, 'proxies'):
3143 proxy_map
.update(handler
.proxies
)
3144 self
._write
_string
('[debug] Proxy map: ' + compat_str(proxy_map
) + '\n')
3146 if self
.params
.get('call_home', False):
3147 ipaddr
= self
.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3148 self
._write
_string
('[debug] Public IP address: %s\n' % ipaddr
)
3150 latest_version
= self
.urlopen(
3151 'https://yt-dl.org/latest/version').read().decode('utf-8')
3152 if version_tuple(latest_version
) > version_tuple(__version__
):
3153 self
.report_warning(
3154 'You are using an outdated version (newest version: %s)! '
3155 'See https://yt-dl.org/update if you need help updating.' %
3158 def _setup_opener(self
):
3159 timeout_val
= self
.params
.get('socket_timeout')
3160 self
._socket
_timeout
= 600 if timeout_val
is None else float(timeout_val
)
3162 opts_cookiefile
= self
.params
.get('cookiefile')
3163 opts_proxy
= self
.params
.get('proxy')
3165 if opts_cookiefile
is None:
3166 self
.cookiejar
= compat_cookiejar
.CookieJar()
3168 opts_cookiefile
= expand_path(opts_cookiefile
)
3169 self
.cookiejar
= YoutubeDLCookieJar(opts_cookiefile
)
3170 if os
.access(opts_cookiefile
, os
.R_OK
):
3171 self
.cookiejar
.load(ignore_discard
=True, ignore_expires
=True)
3173 cookie_processor
= YoutubeDLCookieProcessor(self
.cookiejar
)
3174 if opts_proxy
is not None:
3175 if opts_proxy
== '':
3178 proxies
= {'http': opts_proxy, 'https': opts_proxy}
3180 proxies
= compat_urllib_request
.getproxies()
3181 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3182 if 'http' in proxies
and 'https' not in proxies
:
3183 proxies
['https'] = proxies
['http']
3184 proxy_handler
= PerRequestProxyHandler(proxies
)
3186 debuglevel
= 1 if self
.params
.get('debug_printtraffic') else 0
3187 https_handler
= make_HTTPS_handler(self
.params
, debuglevel
=debuglevel
)
3188 ydlh
= YoutubeDLHandler(self
.params
, debuglevel
=debuglevel
)
3189 redirect_handler
= YoutubeDLRedirectHandler()
3190 data_handler
= compat_urllib_request_DataHandler()
3192 # When passing our own FileHandler instance, build_opener won't add the
3193 # default FileHandler and allows us to disable the file protocol, which
3194 # can be used for malicious purposes (see
3195 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3196 file_handler
= compat_urllib_request
.FileHandler()
3198 def file_open(*args
, **kwargs
):
3199 raise compat_urllib_error
.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3200 file_handler
.file_open
= file_open
3202 opener
= compat_urllib_request
.build_opener(
3203 proxy_handler
, https_handler
, cookie_processor
, ydlh
, redirect_handler
, data_handler
, file_handler
)
3205 # Delete the default user-agent header, which would otherwise apply in
3206 # cases where our custom HTTP handler doesn't come into play
3207 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3208 opener
.addheaders
= []
3209 self
._opener
= opener
3211 def encode(self
, s
):
3212 if isinstance(s
, bytes):
3213 return s
# Already encoded
3216 return s
.encode(self
.get_encoding())
3217 except UnicodeEncodeError as err
:
3218 err
.reason
= err
.reason
+ '. Check your system encoding configuration or use the --encoding option.'
3221 def get_encoding(self
):
3222 encoding
= self
.params
.get('encoding')
3223 if encoding
is None:
3224 encoding
= preferredencoding()
3227 def _write_thumbnails(self
, info_dict
, filename
): # return the extensions
3228 write_all
= self
.params
.get('write_all_thumbnails', False)
3230 if write_all
or self
.params
.get('writethumbnail', False):
3231 thumbnails
= info_dict
.get('thumbnails') or []
3232 multiple
= write_all
and len(thumbnails
) > 1
3235 for t
in thumbnails
[::1 if write_all
else -1]:
3236 thumb_ext
= determine_ext(t
['url'], 'jpg')
3237 suffix
= '%s.' % t
['id'] if multiple
else ''
3238 thumb_display_id
= '%s ' % t
['id'] if multiple
else ''
3239 thumb_filename
= replace_extension(filename
, suffix
+ thumb_ext
, info_dict
.get('ext'))
3241 if not self
.params
.get('overwrites', True) and os
.path
.exists(encodeFilename(thumb_filename
)):
3242 ret
.append(suffix
+ thumb_ext
)
3243 self
.to_screen('[%s] %s: Thumbnail %sis already present' %
3244 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
3246 self
.to_screen('[%s] %s: Downloading thumbnail %s ...' %
3247 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
))
3249 uf
= self
.urlopen(t
['url'])
3250 with open(encodeFilename(thumb_filename
), 'wb') as thumbf
:
3251 shutil
.copyfileobj(uf
, thumbf
)
3252 ret
.append(suffix
+ thumb_ext
)
3253 self
.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3254 (info_dict
['extractor'], info_dict
['id'], thumb_display_id
, thumb_filename
))
3255 t
['filepath'] = thumb_filename
3256 except network_exceptions
as err
:
3257 self
.report_warning('Unable to download thumbnail "%s": %s' %
3258 (t
['url'], error_to_compat_str(err
)))
3259 if ret
and not write_all
: